pkgsrc change: depend on math/py-pytables.
Changes since 0.9.1:
* Delimited file parsing engine rewritten to use a fraction of memory while
being 40%+ faster.
- Much-improved Unicode handling via the encoding option.
- Column filtering (usecols)
- Dtype specification (dtype argument)
- Ability to specify strings to be recognized as True/False
- Ability to yield NumPy record arrays (as_recarray)
- High performance delim_whitespace option
- Decimal format (e.g. European format) specification
- Easier CSV dialect options: escapechar, lineterminator, quotechar, etc.
- More robust handling of many exceptional kinds of files observed in the wild
* API changes
- Deprecated DataFrame BINOP TimeSeries special case behavior
- Altered resample default behavior
- Infinity and negative infinity are no longer treated as NA by isnull and
notnull.
- Methods with the inplace option now all return None instead of the calling
object.
- pandas.merge no longer sorts the group keys (sort=False) by default.
- The default column names for a file with no header have been changed.
- Values like 'Yes' and 'No' are not interpreted as boolean by default.
- The file parsers will not recognize non-string values arising from a
converter function as NA.
- Calling fillna on Series or DataFrame with no arguments is no longer valid
code.
- Series.apply will now operate on a returned value from the applied function.
- New API functions for working with pandas options.
* New features
- Wide DataFrame Printing.
- Updated PyTables Support.
* Enhancements
- added ability to hierarchical keys.
- added mixed-dtype support!
- performance improvments on table writing.
- support for arbitrarily indexed dimensions.
- SparseSeries now has a density property.
* Bug fixes
- added Term method of specifying where conditions.
- del store['df'] now call store.remove('df') for store deletion.
- deleting of consecutive rows is much faster than before.
- in_itemsize parameter can be specified in table creation to force a minimum
size for indexing columns.
- indexing support via create_table_index (requires PyTables >= 2.3)
- appending on a store would fail if the table was not first created via put.
- fixed issue with missing attributes after loading a pickled dataframe.
- minor change to select and remove: require a table ONLY if where is also
provided.
* Compatibility
- 0.10 of HDFStore is backwards compatible for reading tables created
in a prior version of pandas, however, query terms using the prior
(undocumented) methodology are unsupported.
* N Dimensional Panels (Experimental)
Approved by wiz@
Upstream changes:
CHANGES IN R VERSION 2.15.1:
NEW FEATURES:
o source() now uses withVisible() rather than
.Internal(eval.with.vis). This sometimes alters tracebacks
slightly.
o install.packages("pkg_version.tgz") on Mac OS X now has sanity
checks that this is actually a binary package (as people have
tried it with incorrectly named source packages).
o splineDesign() and spline.des() in package splines have a new
option sparse which can be used for efficient construction of a
sparse B-spline design matrix (_via_ Matrix).
o norm() now allows type = "2" (the 'spectral' or 2-norm) as well,
mainly for didactical completeness.
o pmin() and pmax()) now also work when one of the inputs is of
length zero and others are not, returning a zero-length vector,
analogously to, say, +.
o colorRamp() (and hence colorRampPalette()) now also works for the
boundary case of just one color when the ramp is flat.
o qqline() has new optional arguments distribution, probs and
qtype, following the example of lattice's panel.qqmathline().
o .C() gains some protection against the misuse of character vector
arguments. (An all too common error is to pass character(N),
which initializes the elements to "", and then attempt to edit
the strings in-place, sometimes forgetting to terminate them.)
o Calls to the new function globalVariables() in package utils
declare that functions and other objects in a package should be
treated as globally defined, so that CMD check will not note
them.
o print(packageDescription(*)) trims the Collate field by default.
o The included copy of zlib has been updated to version 1.2.7.
o A new option "show.error.locations" has been added. When set to
TRUE, error messages will contain the location of the most recent
call containing source reference information. (Other values are
supported as well; see ?options.)
o The NA warning messages from e.g. pchisq() now report the call to
the closure and not that of the .Internal.
o Added Polish translations by <c5><81>ukasz Daniel.
PERFORMANCE IMPROVEMENTS:
o In package parallel, makeForkCluster() and the multicore-based
functions use native byte-order for serialization (deferred from
2.15.0).
o lm.fit(), lm.wfit(), glm.fit() and lsfit() do less copying of
objects, mainly by using .Call() rather than .Fortran().
o .C() and .Fortran() do less copying: arguments which are raw,
logical, integer, real or complex vectors and are unnamed are not
copied before the call, and (named or not) are not copied after
the call. Lists are no longer copied (they are supposed to be
used read-only in the C code).
o tabulate() makes use of .C(DUP = FALSE) and hence does not copy
bin. (Suggested by Tim Hesterberg.) It also avoids making a
copy of a factor argument bin.
o Other functions (often or always) doing less copying include
cut(), dist(), the complex case of eigen(), hclust(), image(),
kmeans(), loess(), stl() and svd(LINPACK = TRUE).
o There is less copying when using primitive replacement functions
such as names(), attr() and attributes().
DEPRECATED AND DEFUNCT:
o The converters for use with .C() (see ?getCConverterDescriptions)
are deprecated: use the .Call() interface instead. There are no
known examples (they were never fully documented).
UTILITIES:
o For R CMD check, a few people have reported problems with
junctions on Windows (although they were tested on Windows 7, XP
and Server 2008 machines and it is unknown under what
circumstances the problems occur). Setting the environment
variable R_WIN_NO_JUNCTIONS to a non-empty value (e.g. in
~/.R/check.Renviron) will force copies to be used instead.
INSTALLATION:
o R CMD INSTALL with _R_CHECK_INSTALL_DEPENDS_ set to a true value
(as done by R CMD check --as-cran) now restricts the packages
available when lazy-loading as well as when test-loading (since
packages such as ETLUtils and agsemisc had top-level calls to
library() for undeclared packages).
This check is now also available on Windows.
C-LEVEL FACILITIES:
o C entry points mkChar and mkCharCE now check that the length of
the string they are passed does not exceed 2^31-1 bytes: they
used to overflow with unpredictable consequences.
o C entry points R_GetCurrentSrcref and R_GetSrcFilename have been
added to the API to allow debuggers access to the source
references on the stack.
WINDOWS-SPECIFIC CHANGES:
o Windows-specific changes will now be announced in this file
(NEWS). Changes up and including R 2.15.0 remain in the CHANGES
file.
o There are two new environment variables which control the
defaults for command-line options.
If R_WIN_INTERNET2 is set to a non-empty value, it is as if
--internet2 was used.
If R_MAX_MEM_SIZE is set, it gives the default memory limit if
--max-mem-size is not specified: invalid values being ignored.
BUG FIXES:
o lsfit() lost the names from the residuals.
o More cases in which merge() could create a data frame with
duplicate column names now give warnings. Cases where names
specified in by match multiple columns are errors.
o Nonsense uses such as seq(1:50, by = 5) (from package plotrix)
and seq.int(1:50, by = 5) are now errors.
o The residuals in the 5-number summary printed by summary() on an
"lm" object are now explicitly labelled as weighted residuals
when non-constant weights are present. (Wish of PR#14840.)
o tracemem() reported that all objects were copied by .C() or
.Fortran() whereas only some object types were ever copied.
It also reported and marked as copies _some_ transformations such
as rexp(n, x): it no longer does so.
o The plot() method for class "stepfun" only used the optional xval
argument to compute xlim and not the points at which to plot (as
documented). (PR#14864)
o Names containing characters which need to be escaped were not
deparsed properly. (PR#14846)
o Trying to update (recommended) packages in R_HOME/library without
write access is now dealt with more gracefully. Further, such
package updates may be skipped (with a warning), when a newer
installed version is already going to be used from .libPaths().
(PR#14866)
o hclust() is now fast again (as up to end of 2003), with a
different fix for the "median"/"centroid" problem. (PR#4195).
o get_all_vars() failed when the data came entirely from vectors in
the global environment. (PR#14847)
o R CMD check with _R_CHECK_NO_RECOMMENDED_ set to a true value (as
done by the --as-cran option) could issue false errors if there
was an indirect dependency on a recommended package.
o formatC() uses the C entry point str_signif which could write
beyond the length allocated for the output string.
o Missing default argument added to implicit S4 generic for
backsolve(). (PR#14883)
o Some bugs have been fixed in handling load actions that could
fail to export assigned items or generate spurious warnings in
CMD check on loading.
o For tiff(type = "windows"), the numbering of per-page files
except the last was off by one.
o On Windows, loading package stats (which is done for a default
session) would switch line endings on stdout and stderr from CRLF
to LF. This affected Rterm and R CMD BATCH.
o On Windows, the compatibility function x11() had not kept up with
changes to windows(), and issued warnings about bad parameters.
(PR#14880)
o On Windows, the Sys.glob() function did not handle UNC paths as
it was designed to try to do. (PR#14884)
o In package parallel, clusterApply() and similar failed to handle
a (pretty pointless) length-1 argument. (PR#14898)
o Quartz Cocoa display reacted asynchronously to dev.flush() which
means that the redraw could be performed after the plot has been
already modified by subsequent code. The redraw is now done
synchronously in dev.flush() to allow animations without sleep
cycles.
o Source locations reported in traceback() were incorrect when
byte-compiled code was on the stack.
o plogis(x, lower = FALSE, log.p = TRUE) no longer underflows early
for large x (e.g. 800).
o ?Arithmetic's "1 ^ y and y ^ 0 are 1, _always_" now also applies
for integer vectors y.
o X11-based pixmap devices like png(type = "Xlib") were trying to
set the cursor style, which triggered some warnings and hangs.
o Code executed by the built-in HTTP server no longer allows other
HTTP clients to re-enter R until the current worker evaluation
finishes, to prevent cascades.
o The plot() and Axis() methods for class "table" now respect
graphical parameters such as cex.axis. (Reported by Martin
Becker.)
o Under some circumstances package.skeleton() would give out
progress reports that could not be translated and so were
displayed by question marks. Now they are always in English.
(This was seen for CJK locales on Windows, but may have occurred
elsewhere.)
o The evaluator now keeps track of source references outside of
functions, e.g. when source() executes a script.
o The replacement method for window() now works correctly for
multiple time series of class "mts". (PR#14925)
o is.unsorted() gave incorrect results on non-atomic objects such
as data frames. (Reported by Matthew Dowle.)
o The value returned by tools::psnice() for invalid pid values was
not always NA as documented.
o Closing an X11() window while locator() was active could abort
the R process.
o getMethod(f, sig) produced an incorrect error message in some
cases when f was not a string).
o Using a string as a "call" in an error condition with
options(showErrorCalls=TRUE) could cause a segfault. (PR#14931)
o The string "infinity" allowed by C99 was not accepted as a
numerical string value by e.g. scan() and as.character().
(PR#14933)
o In legend(), setting some entries of lwd to NA was inconsistent
(depending on the graphics device) in whether it would suppress
those lines; now it consistently does so. (PR#14926)
o by() failed for a zero-row data frame. (Reported by Weiqiang
Qian)
o Yates correction in chisq.test() could be bigger than the terms
it corrected, previously leading to an infinite test statistic in
some corner cases which are now reported as NaN.
o xgettext() and related functions sometimes returned items that
were not strings for translation. (PR#14935)
o plot(<lm>, which=5) now correctly labels the factor level
combinations for the special case where all h[i,i] are the same.
(PR#14837)
CHANGES IN R VERSION 2.15.0:
SIGNIFICANT USER-VISIBLE CHANGES:
o The behaviour of unlink(recursive = TRUE) for a symbolic link to
a directory has changed: it now removes the link rather than the
directory contents (just as rm -r does).
On Windows it no longer follows reparse points (including
junctions and symbolic links).
NEW FEATURES:
o Environment variable RD2DVI_INPUTENC has been renamed to
RD2PDF_INPUTENC.
o .Deprecated() becomes a bit more flexible, getting an old
argument.
o Even data-only packages without R code need a namespace and so
may need to be installed under R 2.14.0 or later.
o assignInNamespace() has further restrictions on use apart from at
top-level, as its help page has warned. Expect it to be disabled
from programmatic use in the future.
o system() and system2() when capturing output report a non-zero
status in the new "status" attribute.
o kronecker() now has an S4 generic in package methods on which
packages can set methods. It will be invoked by X %x% Y if
either X or Y is an S4 object.
o pdf() accepts forms like file = "|lpr" in the same way as
postscript().
o pdf() accepts file = NULL. This means that the device does NOT
create a PDF file (but it can still be queried, e.g., for font
metric info).
o format() (and hence print()) on "bibentry" objects now uses
options("width") to set the output width.
o legend() gains a text.font argument. (Suggested by Tim Paine,
PR#14719.)
o nchar() and nzchar() no longer accept factors (as integer
vectors). (Wish of PR#6899.)
o summary() behaves slightly differently (or more precisely, its
print() method does). For numeric inputs, the number of NAs is
printed as an integer and not a real. For dates and datetimes,
the number of NAs is included in the printed output (the latter
being the wish of PR#14720).
The "data.frame" method is more consistent with the default
method: in particular it now applies zapsmall() to
numeric/complex summaries.
o The number of items retained with options(warn = 0) can be set by
options(nwarnings=).
o There is a new function assignInMyNamespace() which uses the
namespace of the function it is called from.
o attach() allows the default name for an attached file to be
overridden.
o bxp(), the work horse of boxplot(), now uses a more sensible
default xlim in the case where at is specified differently from
1:n, see the discussion on R-devel, <URL:
https://stat.ethz.ch/pipermail/r-devel/2011-November/062586.html>.
o New function paste0(), an efficient version of paste(*, sep=""),
to be used in many places for more concise (and slightly more
efficient) code.
o Function setClass() in package methods now returns, invisibly, a
generator function for the new class, slightly preferred to
calling new(), as explained on the setClass help page.
o The "dendrogram" method of str() now takes its default for
last.str from option str.dendrogram.last.
o New simple fitted() method for "kmeans" objects.
o The traceback() function can now be called with an integer
argument, to display a current stack trace. (Wish of PR#14770.)
o setGeneric() calls can be simplified when creating a new generic
function by supplying the default method as the def argument.
See ?setGeneric.
o serialize() has a new option xdr = FALSE which will use the
native byte-order for binary serializations. In scenarios where
only little-endian machines are involved (these days, close to
universal) and (un)serialization takes an appreciable amount of
time this may speed up noticeably transferring data between
systems.
o The internal (un)serialization code is faster for long vectors,
particularly with XDR on some platforms. (Based on a suggested
patch by Michael Spiegel.)
o For consistency, circles with zero radius are omitted by points()
and grid.circle(). Previously this was device-dependent, but
they were usually invisible.
o NROW(x) and NCOL(x) now work whenever dim(x) looks appropriate,
e.g., also for more generalized matrices.
o PCRE has been updated to version 8.30.
o The internal R_Srcref variable is now updated before the browser
stops on entering a function. (Suggestion of PR#14818.)
o There are 'bare-bones' functions .colSums(), .rowSums(),
.colMeans() and .rowMeans() for use in programming where ultimate
speed is required.
o The formerly internal function .package_dependencies() from
package tools for calculating (recursive) (reverse) dependencies
on package databases has been renamed to package_dependencies()
and is now exported.
o There is a new function optimHess() to compute the (approximate)
Hessian for an optim() solution if hessian = TRUE was forgotten.
o .filled.contour() is a 'bare-bones' function to add a
filled-contour rectangular plot to an already prepared plot
region.
o The stepping in debugging and single-step browsing modes has
changed slightly: now left braces at the start of the body are
stepped over for if statements as well as for for and while
statements. (Wish of PR#14814.)
o library() no longer warns about a conflict with a function from
package:base if the function has the same code as the base one
but with a different environment. (An example is Matrix::det().)
o When deparsing very large language objects, as.character() now
inserts newlines after each line of approximately 500 bytes,
rather than truncating to the first line.
o New function rWishart() generates Wishart-distributed random
matrices.
o Packages may now specify actions to be taken when the package is
loaded (setLoadActions()).
o options(max.print = Inf) and similar now give an error (instead
of warnings later).
o The "difftime" replacement method of units tries harder to
preserve other attributes of the argument. (Wish of PR#14839.)
o poly(raw = TRUE) no longer requires more unique points than the
degree. (Requested by John Fox.)
PACKAGE parallel:
o There is a new function mcmapply(), a parallel version of
mapply(), and a wrapper mcMap(), a parallel version of Map().
o A default cluster can be registered by the new function
setDefaultCluster(): this will be used by default in functions
such as parLapply().
o clusterMap() has a new argument .scheduling to allow the use of
load-balancing.
o There are new load-balancing functions parLapplyLB() and
parSapplyLB().
o makePSOCKCluster() has a new option useXDR = FALSE which can be
used to avoid byte-shuffling for serialization when all the nodes
are known to be little-endian (or all big-endian).
PACKAGE INSTALLATION:
o Non-ASCII vignettes without a declared encoding are no longer
accepted.
o C/C++ code in packages is now compiled with -NDEBUG to mitigate
against the C/C++ function assert being called in production use.
Developers can turn this off during package development with
PKG_CPPFLAGS = -UNDEBUG.
o R CMD INSTALL has a new option --dsym which on Mac OS X (Darwin)
dumps the symbols alongside the .so file: this is helpful when
debugging with valgrind (and especially when installing packages
into R.framework). [This can also be enabled by setting the
undocumented environment variable PKG_MAKE_DSYM, since R 2.12.0.]
o R CMD INSTALL will test loading under all installed
sub-architectures even for packages without compiled code, unless
the flag --no-multiarch is used. (Pure R packages can do things
which are architecture-dependent: in the case which prompted
this, looking for an icon in a Windows R executable.)
o There is a new option install.packages(type = "both") which tries
source packages if binary packages are not available, on those
platforms where the latter is the default.
o The meaning of install.packages(dependencies = TRUE) has changed:
it now means to install the essential dependencies of the named
packages plus the Suggests, but only the essential dependencies
of dependencies. To get the previous behaviour, specify
dependencies as a character vector.
o R CMD INSTALL --merge-multiarch is now supported on OS X and
other Unix-alikes using multiple sub-architectures.
o R CMD INSTALL --libs-only now by default does a test load on
Unix-alikes as well as on Windows: suppress with --no-test-load.
UTILITIES:
o R CMD check now gives a warning rather than a note if it finds
inefficiently compressed datasets. With bzip2 and xz compression
having been available since R 2.10.0, it only exceptionally makes
sense to not use them.
The environment variable _R_CHECK_COMPACT_DATA2_ is no longer
consulted: the check is always done if _R_CHECK_COMPACT_DATA_ has
a true value (its default).
o Where multiple sub-architectures are to be tested, R CMD check
now runs the examples and tests for all the sub-architectures
even if one fails.
o R CMD check can optionally report timings on various parts of the
check: this is controlled by environment variable
_R_CHECK_TIMINGS_ documented in 'Writing R Extensions'. Timings
(in the style of R CMD BATCH) are given at the foot of the output
files from running each test and the R code in each vignette.
o There are new options for more rigorous testing by R CMD check
selected by environment variables - see the 'Writing R
Extensions' manual.
o R CMD check now warns (rather than notes) on undeclared use of
other packages in examples and tests: increasingly people are
using the metadata in the DESCRIPTION file to compute information
about packages, for example reverse dependencies.
o The defaults for some of the options in R CMD check (described in
the 'R Internals' manual) have changed: checks for unsafe and
.Internal() calls and for partial matching of arguments in R
function calls are now done by default.
o R CMD check has more comprehensive facilities for checking
compiled code and so gives fewer reports on entry points linked
into .so/.dll files from libraries (including C++ and Fortran
runtimes).
Checking compiled code is now done on FreeBSD (as well as the
existing supported platforms of Linux, Mac OS X, Solaris and
Windows).
o R CMD build has more options for --compact-vignettes: see R CMD
build --help.
o R CMD build has a new option --md5 to add an MD5 file (as done by
CRAN): this is used by R CMD INSTALL to check the integrity of
the distribution.
If this option is not specified, any existing (and probably
stale) MD5 file is removed.
DEPRECATED AND DEFUNCT:
o R CMD Rd2dvi is now defunct: use R CMD Rd2pdf.
o Options such --max-nsize, --max-vsize and the function
mem.limits() are now defunct. (Options --min-nsize and
--min-vsize remain available.)
o Use of library.dynam() without specifying all the first three
arguments is now disallowed.
Use of an argument chname in library.dynam() including the
extension .so or .dll (which was never allowed according to the
help page) is defunct. This also applies to
library.dynam.unload() and to useDynLib directives in NAMESPACE
files.
o The internal functions .readRDS() and .saveRDS() are now defunct.
o The off-line help() types "postscript" and "ps" are defunct.
o Sys.putenv(), replaced and deprecated in R 2.5.0, is finally
removed.
o Some functions/objects which have been defunct for five or more
years have been removed completely. These include .Alias(),
La.chol(), La.chol2inv(), La.eigen(), Machine(), Platform(),
Version, codes(), delay(), format.char(), getenv(), httpclient(),
loadURL(), machine(), parse.dcf(), printNoClass(), provide(),
read.table.url(), restart(), scan.url(), symbol.C(), symbol.For()
and unix().
o The ENCODING argument to .C() is deprecated. It was intended to
smooth the transition to multi-byte character strings, but can be
replaced by the use of iconv() in the rare cases where it is
still needed.
INSTALLATION:
o Building with a positive value of --with-valgrind-instrumentation
now also instruments logical, complex and raw vectors.
C-LEVEL FACILITIES:
o Passing R objects other than atomic vectors, functions, lists and
environments to .C() is now deprecated and will give a warning.
Most cases (especially NULL) are actually coding errors. NULL
will be disallowed in future.
.C() now passes a pairlist as a SEXP to the compiled code. This
is as was documented, but pairlists were in reality handled
differently as a legacy from the early days of R.
o call_R and call_S are deprecated. They still exist in the
headers and as entry points, but are no longer documented and
should not be used for new code.
BUG FIXES:
o str(x, width) now obeys its width argument also for function
headers and other objects x where deparse() is applied.
o The convention for x %/% 0L for integer-mode x has been changed
from 0L to NA_integer_. (PR#14754)
o The exportMethods directive in a NAMESPACE file now exports S4
generics as necessary, as the extensions manual said it does.
The manual has also been updated to be a little more informative
on this point.
It is now required that there is an S4 generic (imported or
created in the package) when methods are to be exported.
o Reference methods cannot safely use non-exported entries in the
namespace. We now do not do so, and warn in the documentation.
o The namespace import code was warning when identical S4 generic
functions were imported more than once, but should not (reported
by Brian Ripley, then Martin Morgan).
o merge() is no longer allowed (in some ways) to create a data
frame with duplicate column names (which confused PR#14786).
o Fixes for rendering raster images on X11 and Windows devices when
the x-axis or y-axis scale is reversed.
o getAnywhere() found S3 methods as seen from the utils namespace
and not from the environment from which it was called.
o selectMethod(f, sig) would not return inherited group methods
when caching was off (as it is by default).
o dev.copy2pdf(out.type = "cairo") gave an error. (PR#14827)
o Virtual classes (e.g., class unions) had a NULL prototype even if
that was not a legal subclass. See ?setClassUnion.
o The C prototypes for zdotc and zdotu in R_ext/BLAS.h have been
changed to the more modern style rather than that used by f2c.
(Patch by Berwin Turlach.)
o isGeneric() produced an error for primitives that can not have
methods.
o .C() or .Fortran() had a lack-of-protection error if the
registration information resulted in an argument being coerced to
another type.
o boxplot(x=x, at=at) with non finite elements in x and non integer
at could not generate a warning but failed.
o heatmap(x, symm=TRUE, RowSideColors=*) no longer draws the colors
in reversed order.
o predict(<ar>) was incorrect in the multivariate case, for p >= 2.
o print(x, max=m) is now consistent when x is a "Date"; also the
"reached ... max.print .." messages are now consistently using
single brackets.
o Closed the <li> tag in pages generated by Rd2HTML(). (PR#14841.)
o Axis tick marks could go out of range when a log scale was used.
(PR#14833.)
o Signature objects in methods were not allocated as S4 objects
(caused a problem with trace() reported by Martin Morgan).
Upstream changes:
0.15 9 December 2012
- Lots of internal changes to Ei, li, Zeta, and R functions:
- Native Zeta and R have slightly more accurate results.
- For bignums, use Math::MPFR if possible. MUCH faster.
Also allows extended precision while still being fast.
- Better accuracy for standard bignums.
- All four functions do:
- XS if native input.
- MPFR to whatever accuracy is desired, if Math::MPFR installed.
- BigFloat versions if no MPFR and BigFloat input.
- standard version if no MPFR and not a BigFloat.
- Add tests for primorial, jordan_totient, and divisor_sum.
- Revamp of the random_prime internals. Also fixes some issues with
random n-bit and maurer primes.
- The random prime and primorial functions now will return a Math::BigInt
object if the result is greater than the native size. This includes
loading up the Math::BigInt library if necessary.
Notable changes in FriCAS 1.1.8 (compared to version 1.1.7):
- Improvements of pattern matching integrator, it can now integrate
in terms of Fresnel integrals and better handles integrals in terms
of Si and Ci.
- Better integration of symbolic derivatives.
- Better normalization of Liouvillian functions.
- New package for computing limits using Gruntz algorithm.
- Faster removal of roots from denominators.
- New domains for mutivariate Ore algebras and partial differential
operators.
- New package for noncommutative Groebner bases.
- New domain for univariate power series with arbitrary exponents.
- New special functions: Shi and Chi.
- New package for noncommutative Groebner bases.
- New domain for univariate power series with arbitrary exponents.
- New special functions: Shi and Chi.
- Several aggregates (in particular tables) allow more general
parameter types.
- New domain for hash tables using equality from underlying
domain.
Bug fixes, in particular:
- Fixed problem with gcd failing due to bad reduction.
- Fixed series of 'acot' and Puiseux series of several special functions.
- Fixed wrong factorization of differential operators.
- Fixed build problem on recent Mac OS X.
Notable changes in FriCAS 1.1.7 (compared to version 1.1.6):
- Improved integration in terms of special functions.
- Updated new graphics framework and graph theory package.
- Added routines for numerical evaluation of several special
functions.
- Added modular method for computing polynomial gcd over algebraic extensions.
- Derivatives of fresnelC and fresnelS are changed to agree
with established convention.
- When printing floats groups of digits are now separated by
underscores (previously were separated by spaces).
- Added C code for removing directories, this speeds up full
build and should avoid build problems on Mac OSX.
Bug fixes, in particular:
- Series expansion now handle poles of Gamma.
- Fixed derivatives of meijerG.
Upstream changes:
Release Notes for 0.7.2New Page Edit Page Page History
These are the release notes for SymPy 0.7.2. SymPy 0.7.2 was released on October 16, 2012.
Major Changes
Python 3 support
SymPy now supports Python 3. The officially supported versions are 3.2 and 3.3, but 3.1 should also work in a pinch. The Python 3-compatible tarballs will be provided separately, but it is also possible to download Python 2 code and convert it manually, via the bin/use2to3 utility. See the README for more
PyPy support
All SymPy tests pass in recent nightlies of PyPy, and so it should have full support as of the next version after 1.9.
Combinatorics
A new module called Combinatorics was added which is the result of a successful GSoC project. It attempts to replicate the functionality of Combinatorica and currently has full featured support for Permutations, Subsets, Gray codes and Prufer codes.
In another GSoC project, facilities from computational group theory were added to the combinatorics module, mainly following the book "Handbook of computational group theory". Currently only permutation groups are supported. The main functionalities are: basic properties (orbits, stabilizers, random elements...), the Schreier-Sims algorithm (three implementations, in increasing speed: with Jerrum's filter, incremental, and randomized (Monte Carlo)), backtrack searching for subgroups with certain properties.
Definite Integration
A new module called meijerint was added, which is also the result of a successful GSoC project. It implements a heuristic algorithm for (mainly) definite integration, similar to the one used in Mathematica. The code is automatically called by the standard integrate() function. This new algorithm allows computation of important integral transforms in many interesting cases, so helper functions for Laplace, Fourier and Mellin transforms were added as well.
Random Variables
A new module called stats was added. This introduces a RandomSymbol type which can be used to model uncertainty in expressions.
Matrix Expressions
A new matrix submodule named expressions was added. This introduces a MatrixSymbol type which can be used to describe a matrix without explicitly stating its entries. A new family of expression types were also added: Transpose, Inverse, Trace, and BlockMatrix. ImmutableMatrix was added so that explicitly defined matrices could interact with other SymPy expressions.
Sets
A number of new sets were added including atomic sets like FiniteSet, Reals, Naturals, Integers, UniversalSet as well as compound sets like ProductSet and TransformationSet. Using these building blocks it is possible to build up a great variety of interesting sets.
Classical Mechanics
A physics submodule named machanics was added which assists in formation of equations of motion for constrained multi-body systems. It is the result of 3 GSoC projects. Some nontrivial systems can be solved, and examples are provided.
Quantum Mechanics
Density operator module has been added. The operator can be initialized with generic Kets or Qubits. The Density operator can also work with TensorProducts as arguments. Global methods are also added that compute entropy and fidelity of states. Trace and partial-trace operations can also be performed on these density operators.
To enable partial trace operations a Tr module has been added to the core library. While the functionality should remain same, this module is likely to be relocated to an alternate folder in the future. One can currently also use sympy.core.Tr to work on general trace operations, but this module is what is needed to work on trace and partial-trace operations on any sympy.physics.quantum objects.
The Density operators, Tr and Partial trace functionality was implemented as part of student participation in GSoC 2012
Expanded angular momentum to include coupled-basis states and product-basis states. Operators can also be treated as acting on the coupled basis (default behavior) or on one component of the tensor product states. The methods for coupling and uncoupling these states can work on an arbitrary number of states. Representing, rewriting and applying states and operators between bases has been improved.
Commutative Algebra
A new module agca was started which seeks to support computations in commutative algebra (and eventually algebraic geometry) in the style of Macaulay2 and Singular. Currently there is support for computing Groebner bases of modules over a (generalized) polynomial ring over a field. Based on this, there are algorithms for various standard problems in commutative algebra, e.g., computing intersections of submodules, equality tests in quotient rings, etc....
Plotting Module
A new plotting module has been added which uses Matplotlib as its back-end. The plotting module has functions to plot the following:
2D line plots
2D parametric plots.
2D implicit and region plots.
3D surface plots.
3D parametric surface plots.
3D parametric line plots.
Differential Geometry
Thanks to a GSoC project the beginning of a new module covering the theory of differential geometry was started. It can be imported with sympy.diffgeom. It is based on "Functional Differential Geometry" by Sussman and Wisdom. Currently implemented are scalar, vector and form fields over manifolds as well as covariant and other derivatives.
Upstream changes:
0.14 29 November 2012
- Compilation and test issues:
Fix compilation on NetBSD
Try to fix compilation on Win32 + MSVC
Speed up some testing, helps a lot with Cygwin on slow machines
Speed up a lot of slow PP areas, especially used by test suite
- XS AKS extended from half-word to full-word.
- Add functions:
jordan_totient generalization of Euler Totient
divisor_sum run coderef for every divisor
- Allow environment variables MPU_NO_XS and MPU_NO_GMP to turn off XS and
GMP support respectively if they are defined and equal to 1.
- Lehmer prime count for Pure Perl code, including use in nth_prime.
prime count 10^9 using sieve:
71.9s PP sieve
0.47s XS sieve
prime count 10^9 using Lehmer:
0.70s PP lehmer
0.03s XS lehmer
- Moved bignum Zeta and R to separate file, only loaded when needed.
Helpful to get the big rarely-used tables out of the main loading.
- Quote arguments to Math::Big{Int,Float} in a few places it wasn't.
Math::Big* coerces the input to a signed value if it isn't a string,
which causes us all sorts of grief.
0.13 19 November 2012
- Fix an issue with prime count, and make prime count available as a
standalone program using primesieve.
0.12 17 November 2012
- Add bin/primes.pl and bin/factor.pl
- Add functions:
primorial product of primes <= n
pn_primorial product of first n primes
prime_set_config set config options
RiemannZeta export and make accurate for small reals
is_provable_prime prove primes after BPSW
is_aks_prime prove prime via AKS
- Add 'assume_rh' configuration option (default: false) which can be set
to allow functions to assume the Riemann Hypothesis.
- Use the Schoenfeld bound for Pi(x) (x large) if assume_rh is true.
- valgrind testing
- Use long doubles for math functions.
- Some fixes and speedups for ranged primes().
- In the PP code, use 2 MR bases for more numbers when possible.
- Fixup of racing SQUFOF, and switch to use it in factor().
- Complete rewrite of XS p-1 factor routine, includes second stage.
- bug fix for prime_count on edge of cache.
- prime_count will use Lehmer prime counting algorithm for largish
sizes (above 4 million). This is MUCH faster than sieving.
- nth_prime now uses the fast Lehmer prime count below the lower limit,
then sieves up from there. This makes a big speed difference for inputs
over 10^6 or so -- over 100x faster for 10^9 and up.
Upstream changes:
Changes in Version 1.7-9
o Added ggplot2 interface through autplot() and fortify() methods. Also
support function facet_free().
Changes in Version 1.7-8
o Added rollsum.
o Bugfix in src/lag.c for the case of k > NROW.
* Fix deadlock bug in MPI transforms (thanks to Michael Pippig for the
bug report and patch, and to Graham Dennis for the bug report).
* Use 128-bit ARM NEON instructions instead of 64-bits. This change
appears to speed up even ARM processors with a 64-bit NEON pipe.
* Speed improvements for single-precision AVX.
* Speed up planner on machines without "official" cycle counters, such as ARM.
pandas is an open source, BSD-licensed library providing
high-performance, easy-to-use data structures and data analysis tools
for the Python programming language.
Upstream changes:
Version 2.01 - 24 October 2012 (bug fix release)
* Fixed error in electronmass
* New definition for au
* Fixed DESTDIR support for installation
* Changed configure script for success with Solaris
* Small changes to manual
* Fix error reporting not working for certain types of error
* Add shadow around display
* Remove reference to removed source file
* Replace lex/bison parser with hand-written parser
* Use g_warning not g_error when unable to parse currency file
* Set window icon correctly
* Port to GtkApplication
* Replace menubar with a GMenu
* Only show thousands separators in decimal mode
* Translation updates
(unimpressed by https://bugzilla.gnome.org/show_bug.cgi?id=670098)
Changes in 5.1.23
Bug fixes
* [ES] removed *86 specific optimizations (rep. #2154)
* [ES] renamed RMS per cent error -> relative error (rep. #2192)
* [ES] a workaround for a bug in recent Xorg not releasing grab of popup menus
* [ES] accept -graphtype chart instead of bar. List possible values in the
help output
* [TK] fix building against png-1.5
* [SM] a grace_np fix
* [ES] IsoLatin5.enc was not included in the distribution
math/p5-Math-Prime-Util.
A set of utilities related to prime numbers. These include multiple sieving
methods, is_prime, prime_count, nth_prime, approximations and bounds for
the prime_count and nth prime, next_prime and prev_prime, factoring
utilities, and more.
Packaged for pkgsrc-WIP by Kamel Derouiche.
Harminv is a free program (and accompanying library) to solve
the problem of harmonic inversion, given a discrete-time,
finite-length signal that consists of a sum of finitely-many
sinusoids (possibly exponentially decaying) in a given bandwidth,
it determines the frequencies, decay constants, amplitudes,
and phases of those sinusoids.
Changes:
* 1.0 (2012-07-25)
- added option ``decimal-comma`` for text format
- fixed bug in CSV format
* 0.9 (2012-05-20)
- added CSV format, or more acurately: delimiter-separated values format.
Supports popular delimiters (``TAB ,;|:/``), quoting (``"``)
and escape character (``\``). Non-numeric fields are read as NaNs.
- added Canberra CNF format
User visible changes.
Version 2.00 - 6 June 2012
General changes:
* Unit lists provide conversion to sums of units (e.g. feet and inches).
* Added --round (-r) and --show-factor (-S) options for use with unit lists.
* Added unit lists aliases (e.g. time, usvol).
* A python script, units_cur, can update currency data.
* Units now flushes its output so you can talk to it with a pipe.
* Units now works in UTF-8 if the system supports it.
* Added --locale (-l) option to set the locale.
* English units such as the gallon are defined in all locales, with US
definitions by default except in the en_BG locale. You can force
the desired definition (Great Britain or US) by setting the
environment variable UNITS_ENGLISH to GB or US.
* Revised and extended the documentation, with more examples.
* Added locale mapping to address Windows locale names.
* Updated and revised units database, including UTF-8 definitions and
the 2010 CODATA.
* Fixed parsing bug for "1..2" or "1.2.3". These now give an error.
* Unit names can end with a number in the form of a subscript that
starts with a '_', as in 'X_3'.
Changes for units definition files:
* Changed location and names of unit database, splitting off currency
Files are /usr/local/share/units/{definitions,currency}.units and
the personal units file is $HOME/.units instead of $HOME/units.dat.
(Under windows personal unit file is unitdef.units.)
* Personal units filename can be changed with MYUNITSFILE environment variable.
* Prefixes can be defined in any order: it is no longer necessary to
define longer ones before shorter ones.
* New definitions replace previous ones. Personal units file is read
after the system file rather than before.
* Changed syntax for function definitions. Instead of [in-unit,out-unit]
you must now write units=[in-unit,out-unit]. Use 'units -c' to find
places in your personal data files that need to be updated.
* Add optional domain and range to function definitions with syntax
domain=[min,max] and range=[min,max]. Either min or max can be
omitted to signal infinity.
* Unit list aliases can be defined with !unitlist command.
* Added !var and !varnot and !endvar to include definitions based on
environment variable settings.
* Added !set to set environment variables (if they aren't already set).
* Added !message to display a message
* Data files are in UTF-8, with !utf8 and !endutf8 commands to mark
definitions that require UTF-8 support.
* Improved error handling when reading units.dat.
Version 1.88 - 15 Feb 2010
* Updated units.dat
* Fixed bug involving readline version 4.2
Version 1.87 - 25 Sep 2007
* Units reads ~/.units.dat if it exists before reading the standard
units database.
* Added support for ** as an exponent operator.
* Type 'search text' to list units that contain 'text' in their name.
* Precedence of '*' changed to match precedence of '/'.
(Multiplication by juxtaposition retains a higher precedence than
'/'.) The --oldstar option selects the old precedence and --newstar
forces the new precedence.
Project moved to another developer.
The changes compared to 2.2.1 are:
o visual garbage fixed (curses version)
o made unquoted strings default, added "-q" flag to turn quotes back on
o made attributes menu more user friendly: no more question when
changing a single cell, one menu option optimized away
o added "-H" command line flag to hide row/column headers; in FLTK,
this means you can only change sheets via Ctrl-Shift-PgUp/PgDn
o expanded command line help
The changes compared to 2.2.0 are:
o Help system improved
o Compile fix for MacOS
o Link shared by default. Set option ENABLE_STATIC to statically link FLTK.
o Bugfixes for CSV export
o Bugfixes for console mode
o Added font styles bold and underline
The changes compared to 2.1.0 are:
o UTF-8 support for curses frontend
o various bug fixes
The changes compared to 2.0.2 are:
o switched to CMake as build system
o Win32 compatibility
o usage improvements in the GUI version
o various bug fixes
The changes compared to 2.0.1 are:
o Low-resource builds (leaving out the help system)
o various bug fixes
The changes compared to 2.0 are:
o Fix a possible crash and some minor bugs
o improve FLTK navigation and cell editing (moving around and
selecting cells now works while editing a formula)
o automatically start a string value when typing a letter
o fix build system
The changes compared to 1.09 are:
o Several bug fixes and changes how files are handled
o FLTK user interface
o Built-in Help and About screen (currently only in FLTK)
o License changed to GPL (see mbox file in doc folder for permission)
New Features of SNNS 4.3
Version 4.3 of SNNS features the following improvements and
extensions over the earlier version 4.2:
1. Included patches avaiable from the SNNS-development project.
2. License changed to LGPL v2.
3. Fixed some bugs in the installation configuration files.
1. The mpfr_get_decimal64 function (available only when MPFR has
been configured with --enable-decimal-float) has several bugs,
fixed by the get_decimal64 patch, which also provides testcases.
It corresponds to the following changesets in the 3.1 branch:
8373, 8378, 8380.
2. The ternary value returned by the mpfr_strtofr function can
be wrong under particular conditions (example). This is fixed
by the strtofr-ternary-value patch, which also provides
testcases. It corresponds to the following changeset in the 3.1
branch: 8399.
Note that tests might need the package installed.
Changes in version 1.0.1:
- Switched to automake 1.11.6, see
https://lists.gnu.org/archive/html/automake/2012-07/msg00023.html
- #14669: Fixed extraction of CC from gmp.h
- Fixed case of intermediate zero real or imaginary part in mpc_fma,
found by hydra with GMP_CHECK_RANDOMIZE=1346362345
Changes:
Version 1.1.8
* Fixed bugs in the Fortran interface.
* Updated pkg-config file.
Version 1.1.7
* Fixed wrong derivative calculation for erf function.
Version 1.1.6
* Added support for erorr function (erf).
Version 1.1.5
* Re-licensed under GNU Public License version 3.
09/09/12 - Allow use of editline library in Mathomatic, because somehow
use of GPL libraries in LGPL code is not allowed, and GNU readline
is GPL. I am so confused about this Debian bug#687063:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=687063
Based on the bug report's information sources,
it appears he is saying the truth about this license issue,
so I will have to upload new versions of
Mathomatic with editline instead of readline to Debian.
The only noticeable difference should be it doesn't save
the history between Mathomatic sessions.
I will comply with all that request it,
though most distributions do not include editline,
which has no major licensing restrictions. Readline is still
perfectly usable and good as before, you just have to link
it in yourself.
09/10/12 - Going to have to make a new release already, 2 days after the last
one, so I can upload this readline licensing fix to Debian by
linking with editline. There is no reason for anyone to upgrade
to version 16.0.4, unless you wish to link with editline instead
of readline. The proper code has been added. All you have to do
is have the editline libraries loaded on your system,
and run "make EDITLINE=1" to compile and link Mathomatic with
editline.
Mathomatic version 16.0.4 released Monday 09/10/12.
CHANGES MADE TO MATHOMATIC 16.0.2 TO BRING IT UP TO THE NEXT VERSION:
New command "set load" loads the current set options startup file again,
displaying the startup file as it reads it in.
If the file doesn't exist, or something is wrong, then an error message is
displayed and the set command returns with failure.
Accidently putting "load" in the startup file is now handled correctly.
08/09/12 - Allow "matho-primes all" and the command "list primes all" in
Mathomatic to continually output consecutive prime numbers.
08/12/12 - Fixed any rman errors in the makefile so they won't be ignored.
08/17/12 - rmath and matho no longer set the debug_level or modulus_mode,
so they can be set by the startup options file.
08/19/12 - Allow the repeat prefix on the approximate command, making it
approximate and simplify as much as the calculate command
does. Hopefully someday it will just give a temporary
result, too. Numerical input into the symbolic math library
now uses "repeat approximate" instead of just "approximate"
to approximate the user's input, so the result is the
same as the Mathomatic application's result.
Disallow the same warnings to be repeatedly displayed, if the
current warning is the same as the previous warning.
08/20/12 - Allow directory names as read command arguments, instructing
the read command to change the current directory to the specified
directory. Without any arguments, the read command now does an
"ls" command in Unix/Linux, and "dir" under MS-Windows, listing
the current directory contents. Running Mathomatic with a
directory name argument now conveniently changes directory to
that directory, then gives you the main prompt.
08/28/12 - Made "integer" type variables much more useful. For example,
the following now happens generally:
1-> i^(4*integer)
#1: i^(4*integer)
1-> simplify
#1: 1
1-> i^((4*integer) + 1)
#2: i^((4*integer) + 1)
2-> simplify
#2: i
2-> i^((4*integer) + 2)
#3: i^((4*integer) + 2)
3-> simplify
#3: -1
3-> i^((4*integer) + 3)
#4: i^((4*integer) + 3)
4-> simplify
#4: -1*i
08/29/12 - Displays "Calculating..." whenever autocalc is used now.
08/30/12 - Added ability to set the normal text color.
Still defaults to no color. Use "set color 0" to set the normal
text color to green, as it has been for many years in the past.
08/31/12 - Removed the "set preserve_surds" option, since the approximate
and calculate commands all take care of undoing that.
The code remains, however "set preserve_surds" is no longer
advertised. I have never used it.
Surds are preserved, for accuracy's sake, by default.
09/05/12 - "set" as a null set option works now, so no one will have any
trouble setting-up the set options startup file.
Mathomatic version 16.0.3 released Saturday 09/08/12.
Upstream changes:
3.0604 July 14, 2012
- Correct a misspelling of "weight" in
lib/Statistics/Descriptive/Smoother/Weightedexponential.pm
- Thanks to Wilhelm for the report.
- Update the scripts/tag-release.pl file for Mercurial.
3.0603 May 15, 2012
- Use in_between to compare decimal numbers
- Smoothing tests were failing because of rounding problems
- Thanks to Andreas J. K.nig for reporting it and to
Fabio Ponciroli for fixing it.
3.0602 May 12, 2012
- Correct a typo:
- https://rt.cpan.org/Ticket/Display.html?id=77145
- Thanks to Salvatore Bonaccorso and the Debian Perl Group
for the report.
3.0601 May 11, 2012
- No longer using Test::Exception in the tests.
- It was used by the tests and not specified in
test_requires/build_requires.
- Thanks to hsk@fli-leibniz.de for the report.
3.0600 May 11, 2012
- Add the smoothing functionality.
- Add the following public methods: add_data_with_samples(),
set_smoother(), get_smoothed_data() to the main module.
- Add the lib/Statistics/Descriptive/Smoother.pm and
lib/Statistics/Descriptive/Smoother/Exponential.pm
lib/Statistics/Descriptive/Smoother/Weightedexponential.pm
modules.
- Thanks to Fabio Ponciroli
- Add the scripts/bump-version-number.pl to facilitate bumping the
version number.
3.0500 May 03, 2012
- Add the get_data_without_outliers() and the set_outlier_filter()
methods.
- See https://bitbucket.org/barbasa/perl-statistics-descriptive/overview
- Thanks to Fabio Ponciroli
3.0400 March 01, 2012
- Fix https://rt.cpan.org/Ticket/Display.html?id=74890
- some function should return undef() in list context so they can
be easily assigned to values in hash initialisations.
- thanks to SLAFFAN for a preliminary patch which was modified
by SHLOMIF (the current Statistics-Descriptive maintainer).
3.0300 February 11, 2012
- Now mean() and median() and other routines return undef() if there are
no data.
- Somewhat incompatible change: some methods that returned undef() under
list context now return an empty list (so it will be false).
- it is generally not recommended to call such methods in list context
as they should always be called in scalar context.
- Resolves https://rt.cpan.org/Ticket/Display.html?id=74693
- thanks to Shawn Laffan for the report and the patch.
3.0203 November 17, 2011
- Fix https://rt.cpan.org/Ticket/Display.html?id=72495 .
- percentile should not die and should return undef if there are
no elements in the collection.
changes:
* ternary if-then-else operator added (C++ like)
* new intrinsic binary operators: "&&", "||"
* A new bulkmode allows submitting large arrays as variables
* intrinsic "and", "or" and "xor" operators have been removed
* Implementation for complex numbers removed
* Function atan2 added
-bugfixes, cleanup, other minor improvements
pkgsrc note: shlib major changed
CHANGES MADE TO MATHOMATIC 16.0.1 TO BRING IT UP TO THE NEXT VERSION:
A nice cleanup and documenting of everything, while my mind still works.
07/23/12 - Enabled links in "manual.pdf" (the Mathomatic User Guide
and Command Reference), they were not working before,
but they work great now.
07/24/12 - Allow -a option ("set alternative") with sets alternative
color mode, useful in MS-Windows when using Cygwin with the
MinGW compiled version.
07/27/12 - Two bug fixes today:
Fixed using rlwrap under Cygwin and possibly other places,
when running rmath.
get_yes_no() question asker wasn't working in Cygwin or rmath.
Fixed to always ask the question, even if the input is not a TTY.
07/28/12 - Reading directories and empty files gives an error now.
07/30/12 - Split off changes.txt to changes.txt and changes_series_15.txt.
changes.txt and NEWS now contain only series 16 changes.
Added tests/batman_gnuplot_bug.in to prove there is a plotting bug
in gnuplot itself.
08/01/12 - Removed directive to use large font in the CSS for all Mathomatic
documentation. This is so it can be browsed easily with a mobile
device. The font size should be set by the user.
08/02/12 - A one-sided expression with an equals sign now only sets the
expression equal to zero if autocalc didn't work on it. In the
Symbolic Math Library, or without autocalc enabled, all is the
same. This makes it more likely purely numerical input is only
calculated, even when preceded or followed by an equals sign,
when autocalc is enabled. Reason for this change:
Why would you want to set a constant equal to 0?
Fixed a bunch more error reporting bugs coded into version 16.0.1
of Mathomatic to apply identical operations to both sides of an
equation. All fixed now. Points to the error correctly now, too.
08/04/12 - Major change to Symbolic Math Library. It now works exactly like
the application when it comes to purely numerical input,
approximating and displaying the result, however sign variables
are not expanded and the result is not 100% simplified,
so running "simplify sign" afterwards helps with that.
To revert to the old way, just turn off autocalc, or set
the numerical input equal to some normal variable; then there
will be no automatic approximation nor simplification.
You can tell when an input has been approximated because it
was numerical input, because it will always be preceded with
"answer =".
08/05/12 - m4/degrees.m4 copied with m4/functions.m4, when installing with
"make m4install", thanks to a suggestion by Reini Urban,
maintainer of the Cygwin version.
Mathomatic version 16.0.2 released Monday 08/06/12.
CHANGES MADE TO MATHOMATIC 16.0.0 TO BRING IT UP TO THE NEXT VERSION:
07/06/12 - Confirmed successful compilation and testing with the
Tiny C compiler (tcc). Only needed to set the tcc linker
library directories to the current gcc library directories
to make it work (using the -L option).
Searched globally for the word "simply" and fixed many outdated
texts in the Mathomatic documentation and READMEs, deleting some
wrongly used "simply"s, too.
07/07/12 - Allow "set modulus_mode language", where language is C, Java,
Python, or positive.
07/08/12 - Now leaving the "UNIX" C preprocessor define alone,
when "HANDHELD" is defined. "UNIX" was previously
forced undefined.
07/11/12 - "make test", "make check", and ./t now display the actual
Mathomatic version number being tested.
matho-primes now has a -v (display version number) option,
like Mathomatic does.
Mathomatic now automatically clears out all old
numeric calculations if it runs out of equation spaces,
requiring no action from the user.
07/12/12 - The solve command never needs the "repeat solve" prefix anymore.
The repeat flag is always set for the solve command, so that
it will always do full simplifies when verifying.
Added quadratic formula derivation and proof to
"tests/quadratic.in".
07/17/12 - I came up with a swinging new and easy way to add, subtract,
multiply, divide, modular and integer divide,
and raise to the power of both sides of an equation by any
expression. To add x+1 to both sides of the current
equation, just type "+=x+1" at the main prompt. To divide both
sides by c^2, type "/=c^2". You can add stuff to non-equations
too, this way. Be sure and use the simplify command after this
if needed, because only a small amount of simplification is done
by default, just enough so you can see what is happening.
07/20/12 - Changed floating point to rational floating point conversion
routine (f_to_fraction()) to ignore converting anything with
over 15 digits, for greater accuracy.
This fixes some small accuracy bugs: "factor number 17!" now gives
an error instead of the wrong value.
07/21/12 - Integrate, Laplace, and Numerical Integrate commands now warn
when the current equation is not a properly solved equation.
Mathomatic version 16.0.1 released Sunday 07/22/12.
Maxima 5.28 change log
Backwards-incompatible changes:
* package stats: removed function simple_linear_regression (superseded by
linear_regression)
New items in core:
* new function generalized_lambert_w
* new functions zn_mult_table, zn_power_table
* new functions for Galois fields: gf_set, gf_char, gf_prim, etc.
New items in share:
* package descriptive: new function principal_components
* package descriptive: new histogram style 'density'
* package stats: new function linear_regression
Other changes:
* revise system for building Maxima on MS Windows
* function gamma_incomplete: improve accuracy for complex bigfloats
* function expintegral_e: improved calculation for large imaginary part
Bug fixes:
3539699: limit of atan2
3538167: Wrong result for definite integral
3534858: wrong answer: limit
3533723: abs_integrate causes stack overflow
3530767: integrate changes k[0] --> k(0)
3530272: nthroot, bad error msg
3529992: Shi (sinh integral) wrong branch, integrate inconsistent
3529144: Error integrating exp(-x)*sinh(sqrt(x)) with domain: complex
3526359: gamma_incomplete(1/5,-32.0) not accurate
3526111: float erf (%i) not working
3522750: assume & integrate
3521596: atan2(sqrt(1-u)*(u-1),1); /* hangup */
3517785: Wrong sign in exponential integral
3517034: polarform error on simple case
mailing list 2012-04-09: Loading gentran
mailing list 2012-03-27: bug in net present value
unnumbered: inequality facts being forgotten
unnumbered: limit(erfc(z), z, inf)
unnumbered: bug in animated_gif
unnumbered: Lisp output not readable
unnumbered: bigfloats parsed incorrectly when ibase is not 10
Upstream Changes:
1.16 2012-08-06 Florent Angly <florent.angly@gmail.com>
* Fixed issue introduced in version 1.15 where rand() took no notice of
argument and irand() did (bug #78200, reported by David Morel)
* Migration of test suite to Test::More and Test::Number::Delta
1.15 2012-06-04 Florent Angly <florent.angly@gmail.com>
* Implemented irand() to draw random integers (bug #73298, feature requested
by crew@cs.stanford.edu)
* Fixed build failure on Hurd (bug #74165, reported and patched by Salvatore
Bonaccorso)
1.14 2012-05-27 Florent Angly <florent.angly@gmail.com>
* Fixed compilation problems with nmake on Windows platforms
(Florent Angly, bug #74984)
* Fixed issues with random seed (bug #77343, solution suggested by
Laurent Dami, fix implemented by Florent Angly)
1.13 2012-01-20 Abhijit Menon-Sen <ams@toroid.org>
* New seed generation and retrieval mechanism by Florent Angly (bug
#64640 on rt.cpan.org).
Changes since version 0.9:
- First release as a GNU package
- License change: LGPLv3+ for code, GFDLv1.3+ (with no invariant sections)
for documentation
- 100% of all lines are covered by tests
- Functions renamed:
mpc_mul_2exp to mpc_mul_2ui, mpc_div_2exp to mpc_div_2ui
- 0^0, which returned (NaN,NaN) previously, now returns (1,+0)
- Removed compatibility with K&R compilers, untestable due to lack of
such compilers
- New functions: mpc_log10, mpc_mul_2si, mpc_div_2si
- Speed-ups:
- mpc_fma
- Bug fixes:
- mpc_div and mpc_norm now return a value indicating the effective
rounding direction, as the other functions
- mpc_mul, mpc_sqr and mpc_norm now return correct results even if there
are over- or underflows during the computation
- mpc_asin, mpc_proj, mpc_sqr: Wrong result when input variable has
infinite part and equals output variable is corrected
- mpc_fr_sub: Wrong return value for imaginary part is corrected
variables, and customization should be done via INSTALL_ENV/MAKE_ENV
respectively. This also solves the internationalization issue fixed
in previous commit properly.
Noted by <joerg> on pkgsrc-changes@, many thanks!
Changes from version 3.1.0 to version 3.1.1:
- Improved MPFR manual.
- Test coverage: 96.5% lines of code.
- Bug fixes (see <http://www.mpfr.org/mpfr-3.1.0/#fixed> or ChangeLog file).
Release Log
Networkx-1.7
Release date: 4 July 2012
Highlights
New functions for k-clique community finding, flow hierarchy, union, disjoint union, compose, and intersection operators that work on lists of graphs, and creating the biadjacency matrix of a bipartite graph.
New approximation algorithms for dominating set, edge dominating set, independent set, max clique, and min-weighted vertex cover.
Many bug fixes and other improvements.
For full details of the tickets closed for this release (added features and bug fixes) see: https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.7
API Changes
See Version 1.7 notes and API changes
Summary of important user-visible changes for version 3.6:
---------------------------------------------------------
** The PCRE library is now required to build Octave. If a pre-compiled
package does not exist for your system, you can find PCRE sources
at http://www.pcre.org
** The ARPACK library is no longer distributed with Octave.
If you need the eigs or svds functions you must provide an
external ARPACK through a package manager or by compiling it
yourself. If a pre-compiled package does not exist for your system,
you can find the current ARPACK sources at
http://forge.scilab.org/index.php/p/arpack-ng
** Many of Octave's binary operators (.*, .^, +, -, ...) now perform
automatic broadcasting for array operations which allows you to use
operator notation instead of calling bsxfun or expanding arrays (and
unnecessarily wasting memory) with repmat or similar idioms. For
example, to scale the columns of a matrix by the elements of a row
vector, you may now write
rv .* M
In this expression, the number of elements of rv must match the
number of columns of M. The following operators are affected:
plus + .+
minus - .-
times .*
rdivide ./
ldivide .\
power .^ .**
lt <
le <=
eq ==
gt >
ge >=
ne != ~=
and &
or |
atan2
hypot
max
min
mod
rem
xor
additionally, since the A op= B assignment operators are equivalent
to A = A op B, the following operators are also affected:
+= -= .+= .-= .*= ./= .\= .^= .**= &= |=
See the "Broadcasting" section in the new "Vectorization and Faster
Code Execution" chapter of the manual for more details.
** Octave now features a profiler, thanks to the work of Daniel Kraft
under the Google Summer of Code mentorship program. The manual has
been updated to reflect this addition. The new user-visible
functions are profile, profshow, and profexplore.
** Overhaul of statistical distribution functions
Functions now return "single" outputs for inputs of class "single".
75% reduction in memory usage through use of logical indexing.
Random sample functions now use the same syntax as rand and accept
a comma separated list of dimensions or a dimension vector.
Functions have been made Matlab-compatible with regard to special
cases (probability on boundaries, probabilities for values outside
distribution, etc.). This may cause subtle changes to existing
scripts.
negative binomial function has been extended to real, non-integer
inputs. The discrete_inv function now returns v(1) for 0 instead of
NaN. The nbincdf function has been recoded to use a closed form
solution with betainc.
** strread, textscan, and textread have been completely revamped.
They now support nearly all Matlab functionality including:
* Matlab-compatible whitespace and delimiter defaults
* Matlab-compatible options: 'whitespace', treatasempty', format
string repeat count, user-specified comment style, uneven-length
output arrays, %n and %u conversion specifiers (provisionally)
** All .m string functions have been modified for better performance or
greater Matlab compatibility. Performance gains of 15X-30X have
been demonstrated. Operations on cell array of strings no longer pay
quite as high a penalty as those on 2-D character arrays.
deblank: Now requires character or cellstr input.
strtrim: Now requires character or cellstr input.
No longer trims nulls ("\0") from string for Matlab
compatibility.
strmatch: Follows documentation precisely and ignores trailing spaces
in pattern and in string. Note that this is documented
Matlab behavior but the implementation apparently does
not always follow it.
substr: Now possible to specify a negative LEN option which
extracts to within LEN of the end of the string.
strtok: Now accepts cellstr input.
base2dec, bin2dec, hex2dec:
Now accept cellstr inputs.
dec2base, dec2bin, dec2hex:
Now accept cellstr inputs.
index, rindex:
Now accept 2-D character array input.
strsplit: Now accepts 2-D character array input.
** Geometry functions derived from Qhull (convhull, delaunay, voronoi)
have been revamped. The options passed to the underlying qhull
command have been changed for better results or for Matlab
compatibility.
convhull: Default options are "Qt" for 2D, 3D, 4D inputs
Default options are "Qt Qx" for 5D and higher
delaunay: Default options are "Qt Qbb Qc Qz" for 2D and 3D inputs
Default options are "Qt Qbb Qc Qx" for 4D and higher
voronoi: No default arguments
** Date/Time functions updated. Millisecond support with FFF format
string now supported.
datestr: Numerical formats 21, 22, 29 changed to match Matlab.
Now accepts cellstr input.
** The following warning IDs have been removed:
Octave:associativity-change
Octave:complex-cmp-ops
Octave:empty-list-elements
Octave:fortran-indexing
Octave:precedence-change
** The warning ID Octave:string-concat has been renamed to
Octave:mixed-string-concat.
** Octave now includes the following Matlab-compatible preference
functions:
addpref getpref ispref rmpref setpref
** The following Matlab-compatible handle graphics functions have been
added:
guidata uipanel uitoolbar
guihandles uipushtool uiwait
uicontextmenu uiresume waitfor
uicontrol uitoggletool
The uiXXX functions above are experimental.
Except for uiwait and uiresume, the uiXXX functions are not
supported with the FLTK+OpenGL graphics toolkit.
The gnuplot graphics toolkit does not support any of the uiXXX
functions nor the waitfor function.
** New keyword parfor (parallel for loop) is now recognized as a valid
keyword. Implementation, however, is still mapped to an ordinary
for loop.
** Other new functions added in 3.6.0:
bicg nthargout usejava
is_dq_string narginchk waitbar
is_sq_string python zscore
is_function_handle register_graphics_toolkit
loaded_graphics_toolkits recycle
** Deprecated functions.
The following functions were deprecated in Octave 3.2 and have been
removed from Octave 3.6.
create_set spcholinv splu
dmult spcumprod spmax
iscommand spcumsum spmin
israwcommand spdet spprod
lchol spdiag spqr
loadimage spfind spsum
mark_as_command sphcat spsumsq
mark_as_rawcommand spinv spvcat
spatan2 spkron str2mat
spchol splchol unmark_command
spchol2inv split unmark_rawcommand
The following functions have been deprecated in Octave 3.6 and will
be removed from Octave 3.10 (or whatever version is the second major
release after 3.6):
cut is_duplicate_entry
cor polyderiv
corrcoef shell_cmd
__error_text__ studentize
error_text sylvester_matrix
** The following functions have been modified for Matlab compatibility:
randperm
CHANGES MADE TO MATHOMATIC 15.8.5 TO BRING IT UP TO THE NEXT VERSION:
Code, documentation, and user interface improvements, corrections,
and cleanup. Fixed many possible bugs, some where the wrong level global
expression buffers were being used.
05/26/12 - Push command improved with better responses. Same functionality.
In the source code, tune-up variable integer_coefficients was
renamed to "factor_out_all_numeric_gcds", because Mathomatic always
tries to have integer coefficients this year, but it doesn't always
factor out all numeric GCDs unless factor_out_all_numeric_gcds
is true, or the factor command is used. The default is false,
for more orderly and revealing coefficients.
Of course, much of what Mathomatic does is
try to improve readability and simplicity.
There is no need to set this variable, just use the factor command.
05/27/12 - Removed C declarations for memmove(3), the defaults in
/usr/include/string.h are probably better and what's wanted in
every case. It would be very odd if this didn't work 100%.
05/28/12 - readline history file renamed to "~/.matho_history" from
"~/.mathomatic_history". File name was too long for CygWin.
Tested thoroughly compiling, installing, and running under
the latest CygWin. Works fine, except for rlwrap. "rlwrap -v"
returns with error, a successful return is how I test for its
existence.
Made output redirection work with the "list primes" command.
06/02/12 - Cleanup of Linux, Mac OS X, and Windows binary distributions.
The Windows binary distribution now includes m4 scripts, in case
CygWin is installed, allowing use of m4 Mathomatic in Windows.
Fixed MinGW version to not output two carriage returns at the
end of every line of list command output.
06/03/12 - If compiled with -DSHOW_RESOURCES, will give total CPU usage and
RSS size in the "version status" command. Requires OS support.
Some OSes will show even more information. Uses getrusage(2).
06/04/12 - Allow breaking out of user line-input requests with Control-C.
Still have to hit the Enter key, but the command will be aborted.
06/05/12 - Added "lib/example.c", the simplest example yet of Symbolic Math
Library usage. Compile with "compile.testmain" or practice
compiling it by hand.
The simplify command now returns the number of expressions
simplified, so you can tell if "simplify sign" worked.
The solve command can now require verification, by using the
"verifiable" option, instead of the "verify" option. This causes
unverifiable solves to return with failure, aborting any reads.
Fixed missing code in internal C function free_mem().
I don't think it was used by anyone. A call to free_mem() is
now made on exit, if Mathomatic is compiled with -DVALGRIND, to
check for memory leaks.
06/08/12 - Added polynomial factoring to GCD result of divide command.
It is always handy to know what the factors are of the GCD.
Allow comma (,) at the end of most input lines. A comma now
terminates an expression instead of giving an error. Allow
commas all over the place, where-ever logical, in any
Mathomatic command-line. They are used as separators,
more so than spaces.
06/09/12 - Cleaned up variables command to always allow the count parameter,
and to line up everything with 8 character wide tabs.
Added ability to place the definite integration bounds on the
integrate command-line, just like the nintegrate command.
Added titles to most help command pages.
06/10/12 - The "factor number" command works much nicer now, and allows comma
separators and zero.
Developers should note that to remain the same as past versions,
HTML mode needs to be "set html all" to output HTML at all times
in both the application and the symbolic math library, even when
redirecting output. Now setting all HTML mode with
"make pdfsheet". "set html" only outputs HTML code to standard
output.
06/13/12 - Added warning in "misc/known_bugs.txt" about LLVM/Clang optimizer
failure when compiling Mathomatic with LLVM/Clang instead of gcc.
If you enable any optimization at all, entering (32^.5) and the
like will hang Mathomatic, putting it in an endless loop.
So when compiling Mathomatic with LLVM/Clang, always disable
optimization with "-O0", so that it will then run and pass
all of the tests in 1 second and not be infinitely slower.
Mathomatic will hang during "make test"
if compiled with optimization enabled using LLVM. Mathomatic is
not noticeably slower when compiled without any optimization,
because everything is memmove(3)s and floating point arithmetic.
06/15/12 - Added repeat option to replace command. A handy feature that
lets you try plugging different values into an equation. It
checks if the result is an identity, too.
06/18/12 - The version command now has a "status" option, which behaves
as before, displaying all version and status information.
The version command by itself now only displays the Mathomatic
version number. Running "mathomatic -v" is now a good way of
testing for the existence of Mathomatic on your system, only
outputting the version number to standard output and exiting
successfully.
06/19/12 - Removed the parenthesizing of variable names in all messages.
If the current expression is a non-equation, then prefixing or
suffixing an expression with "=" will add that expression as
the other equation side now, conveniently making it an equation
you can solve.
06/22/12 - Added equation number ranges option to tally command. Type
"tally -" to resume if the current equation hasn't changed. Type
"tally all" to add together all stored expressions as the starting
value. Specifying equation numbers or ranges will silently add
them, then prompt for the next things to add. The average option
now displays the number of entries (count) each time the average
is displayed. When you exit by typing an empty line, the current
total is saved in the next available equation space and made
current, so it can easily resume with "tally -". "-" by itself
always means the current equation.
gnuplot now works with MS-Windows better. Tried running a Windows
gnuplot test from scratch, without Cygwin, and it didn't work.
It should be mostly fixed now. So go ahead and try plotting
in Windows, after downloading and installing gnuplot. Please
complain if any problems.
Fixed a long-running problem with the plot command, by asking the
user questions, only if needed, so that gnuplot will not give an
error if you are multi-expression plotting.
06/23/12 - Moved load_rc() out of main.c so that the Mathomatic startup set
options file can be loaded by the library, if the developer wishes.
Changed a few things so that "set save" and "set no save" will work
if load_rc() is called beforehand.
06/25/12 - The simplify command has been fixed for optimal integer coefficient
factoring results and so "180*(sides-2)" simplification works
nicely, by keeping the result the same as the start by
factoring out rational constants greater than 1 (this is new),
along with less than 1,
if the coefficients remain or become integers.
Many things cleaned up and finished, like the official
documentation, the "code integer" command, and "examples/fact.c".
06/27/12 - Allow an ASCII string after the "set save" command, to save only
that string in ~/.mathomaticrc, so that string, which should be set
options, is for every Mathomatic session to start with.
For example, "set save bold color" will start out Mathomatic in
bold color mode every time. Enter "set no save" to remove.
"set save" by itself saves all of the current set options for every
future session.
Mathomatic version 16.0.0 released Friday 06/29/12.
Based on PR 46507 by Wen Heping.
* let to register egg-info.
* all files in distfile are not DOS style EOL oter than Windows related files,
so remove extract option for ZIP.
GMPY 1.15 is a bug fix release. The following bugs were fixed:
* Reference counting leak in divmod(x,0).
* Fatal crash in remove(x,1).
* Discontinue use of custom memory allocator. (Fixes compatibility with Sage.)
* Allow up to base-62 integer conversion.
Based on PR 46506 by Wen Heping.
* let to register egg-info.
* marked as incompatible with python 2.5, as new features in 1.3.
but not marked as compatible with 3.x, one file will not be compiled well
both 3.1 and 3.2.
Lease Log:
Networkx-1.6
Release date: 20 November 2011
Highlights
New functions for finding articulation points, generating random bipartite
graphs, constructing adjacency matrix representations, forming graph products,
computing assortativity coefficients, measuring subgraph centrality and
communicability, finding k-clique communities, and writing JSON format output.
New examples for drawing with D3 Javascript library, and ordering matrices with
the Cuthill-McKee algorithm.
More memory efficient implementation of current-flow betweenness and new
approximation algorithms for current-flow betweenness and shortest-path
betweenness.
Simplified handling of "weight" attributes for algorithms that use
weights/costs/values. See Version 1.6 notes and API changes.
Updated all code to work with the PyPy Python implementation http://pypy.org
which produces faster performance on many algorithms.
For full details of the tickets closed for this release (added features and bug
fixes) see: https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.6
API Changes
See Version 1.6 notes and API changes:
http://networkx.lanl.gov/reference/api_1.6.html
Networkx-1.5
Release date: 4 June 2011
For full details of the tickets closed for this release see: https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.5
Highlights
New features
* Algorithms for generating and analyzing bipartite graphs
* Maximal independent set algorithm
* Erd?s-Gallai graphical degree sequence test
* Negative edge cycle test
* More memory efficient Dijkstra path length with cutoff parameter
* Weighted clustering coefficient
* Read and write version 1.2 of GEXF reader format
* Neighbor degree correlation that handle subsets of nodes
* In-place node relabeling
* Many `weighted' graph algorithms now take optional parameter to use
specified edge attribute (default=`weight') (ticket 509)
* Test for distance regular graphs
* Fast directed Erd?s-Renyi graph generator
* Fast expected degree graph generator
* Navigable small world generator
* Waxman model generator
* Geographical threshold graph generator
* Karate Club, Florentine Families, and Davis' Women's Club graphs
API Changes
See Version 1.5 notes and API changes
http://networkx.lanl.gov/reference/api_1.5.html
Bug fixes
* Fix edge handling for multigraphs in networkx/graphviz interface
(ticket 507)
* Update networkx/pydot interface for new versions of pydot
(ticket 506), (ticket 535)
* Fix negative cycle handling in Bellman-Ford (ticket 502)
* Write more attributes with GraphML and GML formats (ticket 480)
* Handle white space better in read_edgelist (ticket 513)
* Better parsing of Pajek format files (ticket 524) (ticket 542)
* Isolates functions work with directed graphs (ticket 526)
* Faster conversion to numpy matrices (ticket 529)
* Add graph[`name'] and use properties to access Graph.name (ticket 544)
* Topological sort confused None and 0 (ticket 546)
* GEXF writer mishandled weight=0 (ticket 550)
* Speedup in SciPy version of PageRank (ticket 554)
* Numpy PageRank node order incorrect + speedups (ticket 555)
Networkx-1.4
Release date: 23 January 2011
New features
* k-shell,k-crust,k-corona
* read GraphML files from yEd
* read/write GEXF format files
* find cycles in a directed graph
* DFS and BFS algorithms
* chordal graph functions
* Prim's algorithm for minimum spanning tree
* r-ary tree generator
* rich club coefficient
* NumPy matrix version of Floyd's algorithm for all-pairs shortest path
* read GIS shapefiles
* functions to get and set node and edge attributes
* and more, see https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.4
API Changes
* gnp_random_graph() now takes a directed=True|False keyword instead of
create_using
* gnm_random_graph() now takes a directed=True|False keyword instead of
create_using
Bug fixes
* see https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.4
Networkx-1.3
Release date: 28 August 2010
See: https://networkx.lanl.gov/trac/timeline
New features
* Works with Python versions 2.6, 2.7, 3.1, and 3.2 (but not 2.4 and 2.5).
* Minimum cost flow algorithms
* Bellman-Ford shortest paths
* GraphML reader and writer
* More exception/error types
* Updated many tests to unittest style. Run with:
"import networkx; networkx.test()" (requires nose testing package)
* and more, see https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.3
API Changes
* minimum_spanning_tree() now returns a NetworkX Graph (a tree or forest)
Bug fixes
* see https://networkx.lanl.gov/trac/query?status=closed&group=milestone&milestone=networkx-1.3
The Munkres module provides an implementation of the Munkres
algorithm (also called the Hungarian algorithm or the Kuhn-Munkres
algorithm), useful for solving the Assignment Problem.
Assignment Problem: Let C be an nxn matrix representing the costs
of each of n workers to perform any of n jobs. The assignment
problem is to assign jobs to workers in a way that minimizes the
total cost. Since each worker can perform only one job and each
job can be assigned to only one worker the assignments represent
an independent set of the matrix C.
Contributed to pkgsrc-wip by Jason Bacon.
ARPACK is a collection of Fortran77 subroutines designed to solve large
scale eigenvalue problems.
The package is designed to compute a few eigenvalues and corresponding
eigenvectors of a general n by n matrix A. It is most appropriate for large
sparse or structured matrices A where structured means that a matrix-vector
product w <- Av requires order n rather than the usual order n**2 floating
point operations. This software is based upon an algorithmic variant of the
Arnoldi process called the Implicitly Restarted Arnoldi Method (IRAM). When
the matrix A is symmetric it reduces to a variant of the Lanczos process
called the Implicitly Restarted Lanczos Method (IRLM). These variants may be
viewed as a synthesis of the Arnoldi/Lanczos process with the Implicitly
Shifted QR technique that is suitable for large scale problems. For many
standard problems, a matrix factorization is not required. Only the action
of the matrix on a vector is needed. ARPACK software is capable of solving
large scale symmetric, nonsymmetric, and generalized eigenproblems from
significant application areas. The software is designed to compute a few (k)
eigenvalues with user specified features such as those of largest real part
or largest magnitude. Storage requirements are on the order of n*k locations.
No auxiliary storage is required. A set of Schur basis vectors for the desired
k-dimensional eigen-space is computed which is numerically orthogonal to working
precision. Numerically accurate eigenvectors are available on request.
Important Features:
o Reverse Communication Interface.
o Single and Double Precision Real Arithmetic Versions for Symmetric,
Non-symmetric, Standard or Generalized Problems.
o Single and Double Precision Complex Arithmetic Versions for Standard
or Generalized Problems.
o Routines for Banded Matrices - Standard or Generalized Problems.
o Routines for The Singular Value Decomposition.
o Example driver routines that may be used as templates to implement
numerous Shift-Invert strategies for all problem types, data types
and precision.
Thiss fixes a compilation issue with aligned_allocator, and a typo
in the ParametrizedLine documentation.
2.0.16:
Fix bug in 3x3 tridiagonlisation (and consequently in 3x3 selfadjoint eigen decomposition).
Fix compilation for new gcc 4.6.
Fix performance regression since 2.0.12: in some matrix-vector product, complex matrix expressions were not pre-evaluated.
Fix documentation of Least-Square.
New feature: support for part<SelfAdjoint>.
Fix bug in SparseLU::setOrderingMethod.
FFTW 3.3.2
* Removed an archaic stack-alignment hack that was failing with
gcc-4.7/i386.
* Added stack-alignment hack necessary for gcc on Windows/i386. We
will regret this in ten years (see previous change).
* Fix incompatibility with Intel icc which pretends to be gcc
but does not support quad precision.
* make libfftw{threads,mpi} depend upon libfftw when using libtool;
this is consistent with most other libraries and simplifies the life
of various distributors of GNU/Linux.
FFTW 3.3.1
* Changes since 3.3.1-beta1:
- Reduced planning time in estimate mode for sizes with large
prime factors.
- Added AVX autodetection under Visual Studio. Thanks Carsten
Steger for submitting the necessary code.
- Modern Fortran interface now uses a separate fftw3l.f03 interface
file for the long double interface, which is not supported by
some Fortran compilers. Provided new fftw3q.f03 interface file
to access the quadruple-precision FFTW routines with recent
versions of gcc/gfortran.
* Added support for the NEON extensions to the ARM ISA. (Note to beta
users: an ARM cycle counter is not yet implemented; please contact
fftw@fftw.org if you know how to do it right.)
* MPI code now compiles even if mpicc is a C++ compiler; thanks to
Kyle Spyksma for the bug report.
FFTW 3.3.2
* Removed an archaic stack-alignment hack that was failing with
gcc-4.7/i386.
* Added stack-alignment hack necessary for gcc on Windows/i386. We
will regret this in ten years (see previous change).
* Fix incompatibility with Intel icc which pretends to be gcc
but does not support quad precision.
* make libfftw{threads,mpi} depend upon libfftw when using libtool;
this is consistent with most other libraries and simplifies the life
of various distributors of GNU/Linux.
CHANGES MADE TO MATHOMATIC 15.8.4 TO BRING IT UP TO THE NEXT VERSION:
General cleanup.
05/11/12 - Renamed fact(x) function to factorial(x), because Maxima uses
the function "factorial(x)" and no one uses "fact(x)". Added
factorial(x) as a standard function in rmath. Removed binary
operator name "mod" from rmath, due to having a different meaning
in conventional mathematics. It previously meant the % operator,
which is a programming language construct and not math.
05/14/12 - Improved user interface of divide command.
05/16/12 - Improved introduction to Mathomatic in "doc/manual.html".
05/17/12 - Made more use of C function !isfinite(double) for better
reliability. Checks for infinity and NaN on user input.
05/18/12 - Solve verify now quick simplifies the solve result, so you
don't have to. This will result in better verification and
less chance of oversized expressions.
05/20/12 - Fixed solving bug where it was throwing away absolute values.
Improved the way Mathomatic looks.
05/21/12 - examples/limits.c is now a highly polished program, LGPL licensed.
Compile with "./compile.limits". Tells the integer and float
C data types, with sizes, characteristics, and verification
for the current C compiler (cc).
Display total number of unique solutions stored with
"simplify sign".
Fixed several errors in the simplify command documentation in the
Mathomatic Command Reference.
05/22/12 - Fixed simplify command to simplify (x^2 - 1)^4/(x + 1)^2 properly,
by factoring repeated factor polynomials at the very end.
Fixed bug in divide command, wasn't calculating the polynomial GCD
sometimes, etc, because input was not being expanded. Broken on
12/14/11, fixed today.
05/23/12 - Fixed limit command to return the original expression when the
limit variable is not found or the expression contains no
variables. A warning is given, telling that this is the case.
Thanks to Parag Magunia for indicating that this was a bug.
Previously these cases only returned with an error message.
"tests/limits.in" limit command regression tests checked and
re-added to the main tests. All still works the same as when
these limit command regression tests were removed long ago.
05/24/12 - Cleanup, and remove "examples/c", replaced with
"examples/compile.limits" and "examples/compile.roots".
Mathomatic version 15.8.5 released Friday 05/25/12.
CHANGES MADE TO MATHOMATIC 15.8.3 TO BRING IT UP TO THE NEXT VERSION:
The version command now displays the last main prompt return value, and
the number of allocated equation spaces.
04/29/12 - Added simplifying trig identities to the tests. Discovered
"solve 0" doesn't work anymore, use "solve for 0" instead.
"solve 0" will be fixed and enhanced shortly.
04/30/12 - "solve 0" and "solve verify 0" work now. "solve verify 0"
will solve for zero and tell you if the equation is an identity
or not, while "solve 0" simply solves the current equation for
zero. "solve all verify 0" will verify that all entered equations
are identities.
With the roots command, the "Inverse check" value is not
displayed unless debugging is enabled now.
The iterative calculate command mode now always tells the
number of completed feedback iterations.
It was not apparent before.
Documented and ignore pause command with demo mode now,
instead of html mode.
Demo mode is specified with -d, html mode with -x.
05/02/12 - Mathomatic now proudly displays when verifying a solve operation.
Too many UI improvements to mention, all relatively minor.
05/03/12 - Added "copy select" option to the copy command, which selects
the first created copy, making it the current equation, rather
than not updating the current equation.
05/04/12 - Ran valgrind on the symbolic math library executable testmain,
after running all the tests, no memory
leaks were found at all! I was surprised! Now I get to pat
myself on the back! I will make another release soon.
05/05/12 - "display simple" is now allowed, in addition to "display mixed",
to display simple or mixed fractions as desired, regardless of
the default.
"set finance" option removed, replace with "set fixed" option.
Works similarly, except now you can do "set fixed_point=0" to have
integer-only output.
"set no fixed" or "set fixed -1" turns off fixed-point
mode, returning you to true floating-point mode.
05/06/12 - Added more integer factoring debugging code. It's a shame that
Mathomatic only does double precision floating point arithmetic.
Made "set no autodelete" the default. Numeric expressions will
not be deleted. "set auto" leaves "autodelete" alone now.
05/07/12 - real and imaginary commands now append "_real" or "_imag" to the
solved for variable name now, if a solved equation. This is for
clarification and ease of use.
05/08/12 - Improved "misc/limits.c" and added GNU LGPL license preamble.
Later moved to directory examples.
-d demo mode now allows using the calculate command without
prompting for the values of any of the variables. This is so the
calculate command can be used with the online versions of
Mathomatic.
05/09/12 - Moved all worthy example source code from directory misc to
directory examples, so they can be included in the binary
distributions. Moved and now available are limits.c, roots.c,
and testprimes.
Improved/shortened the eliminate command messages and made the
"using" syntax consistent.
Need "set debug -2" to suppress everything now, including warnings.
"set debug -1" only suppresses helpful messages now, warnings will
get through. Warnings are usually rather important.
Mathomatic version 15.8.4 released Thursday 05/10/12.
CHANGES MADE TO MATHOMATIC 15.8.2 TO BRING IT UP TO THE NEXT VERSION:
Code cleanup.
03/27/12 - Added simplification of "tests/trig.in" and "tests/hypertrig.in"
to the regression tests.
04/02/12 - Added ability to create all non-re-entrant code, so that the Apple
app store won't complain. The line containing "_REENTRANT" in
"includes.h" has been commented out, so nothing special needs to be
done, when compiling for iOS.
04/07/12 - Added ability to make help command text paragraphs all one long
line, instead of always expecting an 80 column or higher display.
Useful for 40 column displays for example, as long as the display
wraps at 40 columns, the output should look OK. Better than it
did, anyways. Tables and such are preserved. See the beginning
of "help.c".
04/10/12 - Fixed bug failing to run gnuplot with Mathomatic compiled with
MINGW. This only affects the MS-Windows version. So if you have
downloaded gnuplot under MS-Windows, the plot command should work
the same as in all other operating systems, now. Be sure and name
it gnuplot.exe and have it in an executable directory in your PATH.
This fixed bug is thanks to Tom Sturgeon for bringing this
error to my attention. This gnuplot fix will be included in the
next release, version 15.8.3, and is now in the development
version. The problem was caused by differences between the Unix
echo command and the Windows echo command.
04/12/12 - Comparing expressions with the compare command will now tell if
one expression is the negation (times -1) of the other.
04/22/12 - Added the HTML tidy command after rman, so that the HTML errors
created by rman are all fixed.
Added and documented matho_clear(3) to Symbolic Math Library,
replaces clear_all(). Simply rename all occurrences of
clear_all() in your code that uses the library with
matho_clear().
04/23/12 - Improved debugging code and error messages.
04/25/12 - The previous autocalc result is erased every time autocalc is
used, unless "set no autodelete" was done. This is done in case
you would like to keep every calculation you made stored in an
equation space. The default is the previous behavior:
"set autodelete", which only keeps one numerical calculation
in memory. "autodelete" only means something when "autocalc" is
on, and since the library has no calculate command, this is not
available in the symbolic math library.
04/27/12 - Cleanup of equation space selecting, shelling out with !, etc.
In the Mathomatic application, everything is now flushed before
user input (with fflush(NULL)). fflush(NULL) is not used at all
anymore in the symbolic math library.
Mathomatic version 15.8.3 released Saturday 04/28/12.
CHANGES MADE TO MATHOMATIC 15.8.1 TO BRING IT UP TO THE NEXT VERSION:
Change of 12/18/11 partially undone. Preventing power collecting
of absolute values doesn't seem necessary anymore, some other
change done recently must have fixed it.
Now power collecting is always done when requested,
fixing a substantial simplification regression that began on 12/18/11.
03/13/12 - Small fix to main makefile. "mandir" and "docdir" no longer depend
on "datadir".
03/14/12 - Small fix to simplification. Any absolute value (such as
|x*y| = ((x*y)^2)^.5) is now not ever power expanded
(to (x^2)^.5*(y^2)^.5 or (x*x*y*y)^.5), so some results are
simpler, like simplifying |x^2-x|.
03/23/12 - A change to the matho script makes GNU m4 no longer quiet
about any warnings or errors, when running m4 Mathomatic.
Mathomatic version 15.8.2 released Saturday 03/24/12.
* Author: Artem Ignatiev <zazubrik@gmail.com>
* remove require and rake altogether
* gem build and rake gem both work fine without those requires,
* and requiring 'rake' broke bundler
* add rake as development dependency
* Somehow it broken rake on my other project
=== 0.7.0 / 07.05.2012
* Author: Artem Ignatiev <zazubrik@gmail.com>
* use both ruby 1.8 and 1.9 compatible way of getting character code when hashing
* Fix syntax for ruby-1.9
* return gemspec so that bundler can find it
When bundler loads gemspec, it evaluates it, and if the return value is
not a gem specification built, refuses to load the gem.
* Testing worksheet protection
