| Age | Commit message (Collapse) | Author | Lines |
|---|
|
Until now we only checked for the existance and write-ability of the build
directory. But we recently discovered that if the specified build-directory
is in a non-POSIX compatible partition (for example NTFS), permissions
can't be modified and this can cause crashs in some programs (in
particular, while building Perl, see [1]). The thing that makes this
problem hard to identify is that on such partitions, `chmod' will still
return 0 (so it was hard to find).
With this commit, a check has been added after the user specifies the
build-directory. If the proposed build directory is not able to handle
permissions as expected, the configure script will not continue and will
let the user know and will ask them for another directory.
Also, the two printed characters at the start of error messages were
changed to `**' (instead of `--'). When everything is good, we'll use `--'
to tell the user that their given directory will be used as the build
directory. And since there are multiple checks now, the final message to
specify a new build directory is now moved to the end and not repeated in
every check.
[1] https://savannah.nongnu.org/support/?110220
|
|
Until now, the primary Maneage URLs were under GitLab, but since we now
have a dedicated URL and Git repository, its better to transfer to this as
soon as possible. Therefore with this commit, throughout Maneage, any place
that Maneage was referenced through GitLab has been corrected.
Please correct your project's remote to point to the new repository at
`git.maneage.org/project.git', and please make sure it follows the
`maneage' branch. There is no more `master' branch on Maneage.
|
|
Tags are not a fixed piece of history (they can easily be moved and not
imported in a different repository), so they are only confusing in the
context of Maneage (where people should branch-off the main project). the
raw commit hashes are a much more robust way to store a precise moment in
history.
Before this commit, I removed all Tags from the main Git repositories of
Maneage and thus removed any mention of Tags with
`README-hacking.md'. Ofcourse, if a project decides to use tags is upto
them, but we won't implement it in the main branch.
|
|
Roberto Baena recently tried building a new project with Maneage and
provided the following suggestions to make it more clear for a new user:
1) In the part where we talk about creating a Git repository, we should
highlight that it must be empty. This is because some (for example Gitlab)
propose to include a `README' file. But if the project is not empty, Git
will not allow pushing to it.
2) The `(can be done later)' comment was removed from the "Delete dummy
parts") to avoid confusion about applying some of them, but not others: if
only some are done, it may cause problems in the build.
|
|
Until now, the message that we printed just before starting to build
software didn't actually print the current directory, but only `pwd'. With
this commit, this is fixed (it uses the `currentdir' variable that is
already found before).
|
|
We recently fixed the problem of TeXLive that hard-codes the year of its
build in its installation directory. But the note on this problem was still
kept in `README-hacking.md'. That part is now removed.
Also, to help in following the checklist, it is now an ordered list.
|
|
Until now, throughout Maneage we were using the old name of "Reproducible
Paper Template". But we have finally decided to use Maneage, so to avoid
confusion, the name has been corrected in `README-hacking.md' and also in
the copyright notices.
Note also that in `README-hacking.md', the main Maneage branch is now
called `maneage', and the main Git remote has been changed to
`https://gitlab.com/maneage/project' (this is a new GitLab Group that I
have setup for all Maneage-related projects). In this repository there is
only one `maneage' branch to avoid complications with the `master' branch
of the projects using Maneage later.
|
|
Of the GCC dynamically linked libraries we need to manually add RPATH to
all and for `libstdc++' we also need to tell it to link with
`libiconv'. Until now, the conditional to check for libstdc++ was not
working and thus libiconv wasn't been added to it.
With this commit the conditional has been corrected and is now
working. Also, to help in reading the logs, an echo statement was added
after every call to PatchELF.
|
|
Until now, when a the raw tarball of some software wasn't usable, I would
put it under my own webpage, or `akhlaghi.org/reproduce-software'. That
same address was also used as a backup server. However, now the project has
a proper name: Maneage. So I changed the directory on my own server to
`akhlaghi.org/maneage-software'.
With this commit, this new address has replaced the old one. But to avoid
crashes in projects that haven't yet merged with the main Maneage branch,
the old `reproduce-software' still works (its actually a symbolic link to
the new directory now).
|
|
In the previous commit, we remove the `-static' flag from building PatchELF
because it wasn't necessary any more. Howver, the comment for the check
still included it and could be confusing. This is corrected with this
commit. Also, we don't need the `good_static_libc' variable (that was only
defined to pass onto PatchELF). This has also been corrected.
|
|
Until a few commits ago, PatchELF was built statically because it was used
to patch `libstdc++' at the end of the GCC building phase, but PatchELF
also depends on `libstdc++', so it would crash. However, recently when
patching the GCC libraries, we don't directly apply Patchelf to the
library, first we copy it to a temporary place, do the patching, then put
it in its proper place. So the problem above won't happen any more.
With this commit, I am thus removing the static flag from patchelf and
letting it built dynamically all the time. The main problem was that some
systems don't have a static C++ library, so PatchELF couldn't be built
statically. Instead of adding more checks, we just fixed the core
foundation of the problem.
|
|
Until now the software configuration parameters were defined under the
`reproduce/software/config/installation/' directory. This was because the
configuration parameters of analysis software (for example Gnuastro's
configurations) were placed under there too. But this was terribly
confusing, because the run-time options of programs falls under the
"analysis" phase of the project.
With this commit, the Gnuastro configuration files have been moved under
the new `reproduce/analysis/config/gnuastro' directory and the software
configuration files are directly under `reproduce/software/config'. A clean
build was done with this change and it didn't crash, but it may cause
crashes in derived projects, so after merging with Maneage, please
re-configure your project to see if anything has been missed. Please let us
know if there is a problem.
|
|
In the previous commit, I removed the year from the basic installation of
TeXLive packages, but I forgot to correct this in the high-level TeXLive
packages! This is corrected with this commit.
|
|
It is this time of year again: TeXLive has transitioned to its 2020 release
and the year is imprinted into the installation directory of TeXLive. Until
now, we have had to manually change this year and it caused complications
and was very annoying.
With this commit, the explicit year has been removed from TeXLive's
installation and we now simply put a `maneage' instead of the year. I tried
this on another system and it worked nicely. Until the time that we can
fully install LaTeX packages from source tarballs, this is the best thing
we could do for now.
|
|
Elham Saremi recently reported the following errors when building Numpy in
numpy/core/src/npysort/radixsort.c.src: "error: 'for' loop initial
declarations are only allowed in C99 or C11 mode". After some searching, I
found Issue 14147[1] on Numpy's main repository about the same problem. As
described there, apparently Numpy needs C99 compiler, but doesn't check for
it or set it manually (for some strange reason, leaving it to the packagers
to check if they want!!!).
Any way, after a check with Elham, we were able to fix it by adding the
`--std=c99' to CFLAGS of Numpy's build and with this commit, it is now
being implemented in the core Maneage to not cause a problem in any other
project.
[1] https://github.com/numpy/numpy/issues/14147
|
|
Until now, Astropy was instructed to build its own internal copy of the
Expat library. However, with the recent commits before, Maneage now
includes an installation of Expat and Astropy can't keep the two (its
internal version and the project's version) separate, so they conflict and
don't let Astropy get built.
With this commit, the problem is fixed by setting the Expat library as an
explicit dependency of Astropy and asking Astropy to ignore its internal
copy.
While doing this, I recognized that it is much easier and elegant to add
steps in various stages of the `pybuild' function through hooks instead of
variables. So the fifth argument of the `pybuild' function was removed and
now it actually checks if hooks are defined as functions and if so, they
will be called.
The `pyhook_after' function was also implemented in the installation of
`pybind11' (which needed it, given that the 5th argument of `pybuild' was
removed) and after doing a test-build, I noticed that two lines were not
ending with a `\' in `boost' (a dependency of `pybind11').
Commit written originally by Mohammad Akghlaghi
|
|
Until now we would simply return the version numbers as they were written
into the separate files and situations can happen where the version numbers
contain an underscore (`_'). However, this character is a methematical
character in LaTeX, causing LaTeX to complain and abort.
With this commit, a step has been added at the end of the configure script
to convert any possible `_' to `\_'. Once it is commented (a backslash is
put behind it), the underscore will be printed as it is in the final PDF.
This commit was originally written by Mohammad Akhlaghi
|
|
Until now, the M4 that was built on macOS had internal problems (as
discussed in #1): it would simply print `Abort trap: 6' in the output and
abort. After looking at the build of Homebrew, I noticed that they apply a
patch (correct one line) to fix this problem. To be able to apply that
patch on macOS systems, I had fully open up the build recipe of M4 and
atleast on the testing system, it was built successfully.
Also, after successfully building M4, and thus Autoconf and thus Minizip,
we were able to build XLSX I/O on a macOS and found out that the internal
library's full address wasn't being put in the libraries and
executables. With this commit, we now use macOS's `install_name_tool' to
correct the positions of the two `libxlsxio_*' libraries in all its
executables.
This commit was originally written by Mohammad Akhlaghi
|
|
Until this commit, only the word `Minizip' was written into the Minizip
installation target (without the version number of Minizip). With this
commit, this minor bug has been fixed by using the appropiate Make
variable: `$(minizip-version)'.
|
|
Minizip is a dependency of XLSX I/O and until now, I was just using the
most recent version I found (2.9.2), but XLSX I/O is written for the
Minizip 1.x series, not 2.x. Somehow it didn't cause a crash on my
computer!!! I think XLSX I/O's CMake is instructed to look into system
directories by default when it doesn't find the directories in the given
places. And because I had installed Minizip on my operating system, it
did't complain.
Upon trying the build on their systems, Yahya, Raul and Zahra reported a
failure in the build of XLSX I/O which was due the to the problem above (we
were installing the wrong version of Minizip!).
With this commit, this has been fixed by installing the 1.x series of
Minizip (whish is actually installed within zlib!).
This commit was original done by Mohammad Akhlaghi.
|
|
With this commit, CMake has been updated to its most recent version.
This upgrade has been done because in the installation of XLSX I/O on
macOS laptop, it crashes complaining about C compiler "not able to
compile a simple test program". After a fast search, I found it could be
possible to just use the most recent version of CMake to solve the
problem. But it didn't work. In any case, it is good to have the most
recent version of CMake included.
|
|
There weren't any conflicts in this merge.
|
|
The subdirectories here (and the fact that they may be symbolic links) may
be confusing for some early project users, so a `README.md' file was added
there describing them and when they are links, when directories and when
some may not yet exist.
|
|
Until this commit, the year of this paper was 2019 and the linking url
was the temporal one. However, the final official publication year is
2020. With this commit, the year and the url have been changed to the
final ones.
|
|
With this commit, multiples tabs in the definition of MissFITS tarball
have been removed. Now they are white spaces.
|
|
Until this commit, PyYAML was not set as prerequisite of Astropy. This
package is an optional dependency of Astropy for some particular
functions. However, we have already included PyYAML into this project so
it is available. With this commit, PyYAML has been set as a prerequisite
of Astropy.
In addition to this, Html5lib and Beautifulsoup4 have been also added as
prerequsites of Astropy (and removed from Astroquery prerequisites). I
noticed that both of them are optional dependencies of Astropy.
|
|
In the last update of Astropy to version 4.0 they removed some things
that the previous version of Astroquery needs. As a consequence, it is
also necessary to update the Astroquery version to be a ble to run with
the Astropy 4.0. With this commit, the update of Astroquery to it most
recent version (0.4) has been done.
|
|
Until now, the preparation phase was always executed before the final build
phase when running `./project make'. But when it becomes necessary, project
preparation can be slow and will un-necessarily slow down the project while
the project is growing (focus is on the analysis that is done after
preparation).
With this commit, preparation will be done automatically the first time
that the project is run (`.build/software/preparation-done.mk' doesn't
exist). However, after preperation is complete once, future runs of
`./project make' won't do preparation any more (by calling
`top-prepare.mk'). They will directly call `top-make.mk' for the analysis.
To manually invoke preparation after the first attempt, the `./project
make' script should be run with the new `--prepare-redo' option.
Also, since the preparation phase is now automatically done before the
analysis phase, the long notice that describes running `./project make' at
the end of the preparation phase has been removed in `top-prepare.mk'. It
now just prints a short line, saying the preparation has been complete.
Finally, when the project has not been run with the proper group
configuration, it ends with an `exit 1' so the main `./project' script
doesn't proceed any further.
|
|
MissFITS is package for manipulating FITS files.
I added it as my first commit to the project for educational
purposes.
|
|
Until now, the final preparation target of the preparation phase depended
on all the `$(makesrc)' files. This caused a problem because we were
telling it to also depend on `prepare.tex' (which is the same file that is
being built).
With this commit, we are applying the same solution we have already done in
`paper.mk' (for `paper.tex'): we are removing `prepare' from the list of
prerequisites.
This bug was found by Zahra Sharbaf.
|
|
Until now, like all software on GNU/Linux systems GCC would be built in RAM
(to speed up the build slightly and also not put too much stress on the
HDDs/SSDs). But some systems don't have enough RAM for building GCC and
will complain and crash.
With this commit, we have added a check on the amount of free space in the
`build_tmp' directory (which will be `/dev/shm' on GNU/Linux systems). If
the amount of free space isn't more than 10GB, then GCC won't be built
there and a temporary directory will be built under the `$(BDIR)/software'
directory for it.
This bug was found by Zahra Sharbaf.
This fixes bug #57853.
|
|
I had forgot to add a `./' before the call to `project' for the
`--check-config'.
|
|
XLSX I/O is a very simple and fast program and library for reading and
writing `.xls' and `.xlsx' files (mainly used by Microsoft Excel) to CSV
files. It has two separate executables that can be called for an Excel file
and will output a CSV plain text file that can then be used within the
pipeline with more standard tools.
|
|
Until now we were mistakenly reporting the version of SED instead of
Texinfo.
With this commit, we corrected it!
This was reported by Raul Infante Sainz.
|
|
Until now, we defined `LIBRARY_PATH' to fix the problem of the `ld' linker
of Binutils needing several `*crt*.o' files to run. However, some software
(for example ImageMagick) over-write `LIBRARY_PATH', therefore there is no
other way than to put a link to these necessary files in our local build
directory.
With this commit, we fixed the problem by putting a link to the system's
relevant files in the local library directory. This fixed the problem with
ImageMagick. Later, when we build the GNU C Library in the project, we
should remove this step.
This bug reported by Raul Castellanos Sanchez.
|
|
Until now, when a Fortran compiler didn't exist on the host operating
system, the configure script would crash with a warning. But some projects
may not need Fortran, so this is just an extra/annoying crash!
With this commit, it will still print the warning, but instead of a crash,
it will just sleep for some seconds, then continue. Later, when if a
software needs Fortran, it's building will crash, but atleast the user was
warned.
In the future, we should add a step to check on the necessary software and
see if Fortran is necessary for the project or not. The project
configuration should indeed crash if Fortran is necessary, but we should
tell the user that software XXXX needs Fortran so we can't continue without
a Fortran compiler.
Also, a small sentence ("Project's configuration will continue in XXXX
seconds.") was added after all the warnings that won't cause a crash, so
user's don't think its a crash.
|
|
Until now, the main download script could only check one server for the
given URL. However, ultimately the actual server that a file is downloaded
from is irrelevant for this project: we actually check its
checksum. Especially in the case of software (which are distributed over
many servers), this can usually be very annoying: the servers may not
properly communicate with the running system and even the 10 trials won't
be enough.
With this commit, the download script
`reproduce/analysis/bash/download-multi-try' can take a new optional
argument (a 5th argument). It assumes this argument is a space-separated
list of server(s) to use as backup for the original URL. When downloading
from the original URL fails, it will look into this list and try
downloading the same file from each given server.
|
|
There were no conflicts in this merge.
|
|
Until now, when `./project make' was run after an insuccessful run of
`./project configure', it would just say to run `./project configure'. But
for a first time user, this could be confusing because when the
configuration is done in parallel, the error message can be very high on
the command-line outputs and not seen clearly.
With this commit, the error message is more complete and describes the
problem and what the users should do in which circumstance.
|
|
Until now, Make was just run ordinarily on the two Makefiles of the
software building phase. Therefore when there was a problem with one
software while building in parallel, Make would only complete the running
rules and stop afterwards. But when other rules don't depened on the
crashed rule, its a waste of time to stop the whole thing.
With this commit, both calls to Make in the `configure.sh' script are done
with the `-k' option (or `--keep-going' in GNU Make). With this option, if
a rule crashes, the other rules that don't depend on it will also be
run. Generally, anything that doesn't depend on the crashed rule will be
done. The `-k' option is a POSIX definition in Make, so it is present in
most implemenetations (for the call to `basic.mk').
|
|
Until now the shell scripts in the software building phase were in the
`reproduce/software/bash' directory. But given our recent change to a
POSIX-only start, the `configure.sh' shell script (which is the main
component of this directory) is no longer written with Bash.
With this commit, to fix that problem, that directory's name has been
changed to `reproduce/software/shell'.
|
|
Until now, the project would first ask for the basic directories, then it
would start testing the compiler. But that was problematic because the
build directory can come from a previous setting (with `./project configure
-e'). Also, it could confuse users to first ask for details, then suddently
tell them that you don't have a working C library! We also need to store
the CPATH variable in the `LOCAL.conf' because in some cases, the compiler
won't work without it.
With this commit, the compiler checking has been moved at the start of the
configure script. Instead of putting the test program in the build
directory, we now make a temporary hidden directory in the source directory
and delete that directory as soon as the tests are done.
In the process, I also noticed that the copyright year of the two hidden
files weren't updated and corrected them.
|
|
Rencely the building of GCC was allowed on Debian-based systems that have
their basic C library in architecture-specific directories, like
`/usr/lib/x86_64-linux-gnu'. However, these systems also have their headers
in non-standard locations, for example `/usr/include/x86_64-linux-gnu' and
this caused a crash on a new Ubuntu system.
/usr/include/stdio.h:27:10: fatal error: bits/libc-header-start.h: No
such file or directory
27 | #include <bits/libc-header-start.h>
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
The reason it didn't cause problems on other Ubuntus that we tested before
was historic: In the old days, we would ask Ubuntu systems to install
multilib features to have GCC. Because they had installed those features,
this problem didn't show up! But this wasn't mandatory!
With this commit, the `CPATH' environment variable is set (similar to how
`LIBRARY_PATH' was set) and this fixed the problem on a clean Debian
virtual machine.
This bug was reported by Sebastian Luna Valero.
|
|
Until this commit, the small section of `README-hacking.md' in which it
is explained how to do the first configuration of Git was at the
beginning of the section `First custom commit'. However, it is better to
have it just before the item `Your first commit' in that section. With
this commit, this change has been done. Now the reader has the necessary
steps for configuring Git just before it is needed for making the first
commit.
|
|
Until now, the initial project scripts were primarily tested with GNU
Bash. But Bash is not generally available on all systems (it has many
features beyond POSIX). Because of this, effectively we were imposing the
requirement on the user that they must have Bash installed. We recently
started this with setting the shebang of `project' and
`reproduce/software/bash/configure.sh' to `/bin/sh'. After doing so, Raul
and Gaspar reported an error on their systems.
To fix the problem, I installed Dash (a minimalist POSIX-compliant shell)
on my computer and temporarily set the shebangs to `/bin/dash', ran the
project configuration step and fixed all issues that came up. With this
commit, it can go all the way to building GCC on my system's Dash. After
this stage (when `high-level.mk' is called), there is no problem, because
we have our own version of GNU Bash and that installed version is used.
Probably some more issues still remain and will hopefully be found in the
future.
While doing this, I also noticed the following two minor issues:
- The `./project configure' option `--input-dir' was not recognized
because it was mistakenly checking `--inputdir'. It has been corrected.
- The test C programs now use the `<<EOF' method instead of `echo'.
- In `basic.mk', the extra space between `syspath' and `:=' was removed
(it was an ancient relic!).
|
|
Until now, the hashbang of these two shell scripts was set to `/bin/bash',
hence assuming that GNU Bash exists on the host system! But this is an
extra requirement on the host operating system and these two scripts should
be written such that they operate on a POSIX shell (the generic `/bin/sh'
which can point to any shell program).
With this commit this has been implemented! We may confront some errors as
the system is run on other systems, but we should fix such errors and work
hard to make these two scripts as POSIX-compatible as possible (runnable on
any shell, so as not to force users to install Bash before running the
project).
This completes Task #15525.
|
|
Until now, the main commands to run the project were these: `./project
configure' (to build the software), `./project prepare' (to possibly
arrange input datasets and build special configuration Makefiles) and
finally `./project make' to run the project.
The main logic behind the "prepare" phase `top-prepare.mk' is to build
configuration files that can be fed into the "make" step and optimize its
operation. For example when the total number of necessary inputs for the
majority of the analysis is not as large as the total number of
inputs. With "prepare" (when necessary), you go through the raw inputs,
select the ones that are necessary for the rest of the project. The output
of `top-prepare.mk' is a configuration file (a Make variable) that keeps
the IDs (numbers, names, etc). That configuration file would then be used
in the `top-make.mk' to identify the lower level targets and allow optimal
project organization and management.
But the last two are both part of the analysis, and while they indeed need
different calls to Make to be executed, many projects don't actually need a
preparation phase: ultimately, its an implementation choice by the project
developers and doesn't concern the project users (or the developers when
they are running it).
To avoid confusing the users, or simply annoying them when a projet doesn't
need it, with this commit, the top-level `top-prepare.mk' and `top-make.mk'
Makefiles are called with the single `./project make' command and
`./project prepare' has been dropped. I noticed this while writing the
paper on this system.
|
|
Since I (Raul) did some changes (and I hope to do more :-)) in the
`README-hacking.md', I am adding my information at the beginning of this
file.
|
|
Until this commit, we were asuming that Git was already properly
configured. However, in order to be as complete as possible, it would
be good if the basic commands to configure Git were in the
`README-hacking.md'. With this commit, a small paragraph has been added
in order to have the basic Git configuration commands (i.e. to configure
the name, email, and favorite text editor).
|
|
Until now, the explanation for a missing static C library didn't actually
guide the users to look above and see the error message! So with this
commit, I edited it a little to be more clear (and mention to look
above). Also, I noticed that on Amazon AWS systems, the static C library is
installed as a separate package, so to help the users, I added the
necessary command and some better explanation.
|
