summaryrefslogtreecommitdiff
path: root/pkgtools
AgeCommit message (Collapse)AuthorFilesLines
2005-09-01Added an (undocumented) option -r/--recursive that checks all the packagesrillig1-9/+14
in a category Makefile that are not commented out.
2005-09-01fix typo in comment, reported by Leonard Schmidt on tech-pkg@hubertf1-1/+1
2005-08-25Update to 20050817.xtraeme2-6/+6
Changes: * Patches to make this build on NetBSD -current * Various contribs plus colorization and fixes
2005-08-25Added a check for PLISTs: Man pages should be installed into man/, notrillig1-1/+4
share/man/.
2005-08-25Removed all references to the -B/contblank command line option.rillig2-12/+4
2005-08-24Removed the -p flag from the usage message. It hasn't been implementedrillig1-2/+1
for quite some time. I don't know when it had been removed, but apparently no-one had used it anyway.
2005-08-24Removed the check for contiguous blank lines completely. As long asrillig1-11/+1
pkglint does not catch the important facts correctly it shouldn't be nitpicking about white-space. I doubt that anyone has used the -B# and the -Wwhitespace options at all, so there's no harm removing them.
2005-08-24Added a -Wwhitespace flag that is off by default to control whetherrillig1-2/+4
multiple contiguous blank lines should result in warnings.
2005-08-23Update to 1.22:jmmv3-8/+15
- Honour INSTALL_PACKAGES during the 'auto' target. Found by jwise@.
2005-08-22Sorted the regular expression constants alphabetically. Renamed two of them.rillig1-7/+7
2005-08-22Don't warn about unsorted PLIST entries that contain ${foo}. Only makerillig1-33/+17
sure the others are sorted.
2005-08-21Reduced the number of warnings for replacing ":=" with "=". Added a commandrillig1-3/+10
line flag -Winternal to enable checking of the pkgsrc infrastructure files. By default these files are not checked.
2005-08-21Buildlink files may use the := assignment operator for variable names thatrillig1-2/+2
contain "BUILDLINK". All other uses are still flagged.
2005-08-21Added a flag -g or --gcc-output-format that makes the messages parseable byrillig1-29/+51
the Emacs compile.el program. Changed the keyword for errors from FATAL to ERROR. Sorted command line options alphabetically in the source code.
2005-08-21This is pkglint-4.20. Changes since 4.19 include:rillig1-2/+2
- Very much code cleanup - Introduction of the --autofix flag for category Makefiles - Rewrite of the code to check the DISTNAME section in package Makefiles - Rewrite of the code to check category Makefiles
2005-08-21Generally warn about the use of the := operator in Makefiles.rillig1-2/+10
2005-08-21Rewrote the checking of the Makefile variables in the DISTNAME section.rillig1-74/+67
This reduces the number of false warnings.
2005-08-21include digest-types.h so the various {u,}intNN_t types are defined.grant1-0/+2
fixes build on BSD/OS. from T. M. Pederson in PR pkg/31024.
2005-08-20Use REQD_DIRS instead of MAKE_DIRS for directories under ${PREFIX} sojlam1-3/+4
that these directories will be conditionally removed (based on reference counts), regardless of the value of PKG_CONFIG. Bump the PKGREVISION for packages that were modified as a result.
2005-08-20Fixed a syntax error (a string had not been enclosed in double quotes).rillig1-2/+2
2005-08-20The default values for some variables in checkfile_package_Makefile arerillig1-24/+19
chosen more carefully to avoid false warnings. The local variables are sorted by categories. Unnecessary variables have been removed.
2005-08-19Added the command line option -CDESCR, for consistency.rillig1-2/+4
2005-08-19The --help message was wrong. The progress messages are not printed onrillig1-2/+2
STDERR but on STDOUT.
2005-08-19Reduced two errors to warnings for the category Makefiles: The sortrillig1-3/+3
order and subdirectories which are commented out without giving a reason.
2005-08-19Don't check the distinfo twice if it is defined as ./distinfo in therillig1-2/+2
package Makefile. This occurs for example in multimedia/gmplayer.
2005-08-17Changed the way how the direct use of tools is detected in Makefiles.rillig1-14/+20
Now the valid sequences are deleted from the string. If after the deletion the string contains a tool name, a warning is printed.
2005-08-17Moved the checkdir_package subroutine directly before the main subroutinerillig1-97/+86
to reduce the number of necessary forward declarations. Subsequently, removed those forward declarations.
2005-08-17Renamed check_{package,category} to checkdir_{package,category} torillig1-7/+7
distinguish them from the checkfile_* and checklines_* subroutines.
2005-08-17Fixed some bugs that appeared while checking the databases/ categoryrillig1-3/+4
Makefile.
2005-08-17When checking category Makefiles, avoid false warnings aboutrillig1-40/+63
subdirectories that are in one of filesystem/Makefile, but not in the other.
2005-08-17Don't require the second line of a category Makefile to contain an emptyrillig1-72/+117
comment. Instead, allow arbitrary comments at the beginning of the file. This change required a major rewrite of the code, which had been quite inflexible before.
2005-08-08Don't install the manpage on FreeBSD.xtraeme1-2/+3
2005-08-07Added an experimental --autofix or -F command line option. It currentlyrillig1-12/+89
has only effects for category Makefiles. While implementing it I probably fixed some other bugs in the category Makefile checking code.
2005-08-06Fixed a bug which led to an endless loop. While happily changing therillig1-3/+2
code, I had forgotten one assumption I made. It was that in the ifelse chain the case ( && ) couldn't occur because the outer loop already cought that. And then I had inserted code that violated exactly this assumption. Fixed.
2005-08-06Fixed the indentation. Replaced a nested if with an &&. Noticed that Irillig1-21/+17
had put the closing indentation FIXME comment at a wrong place. But as it is removed now, this doesn't matter.
2005-08-06The check_category subroutine has been converted to a one-pass checker.rillig1-34/+57
The new code, which is not as nice as the old code, will hopefully allow the --autofix option to be implemented really easily. If that will not become true, I will probably revert to the old code. There's an indentation FIXME in the code to keep this patch small. I will remove that FIXME in a minute.
2005-08-06Don't be too strict to the "wip" category Makefile. It may have arillig1-2/+11
different second line and additional targets after the SUBDIR entries.
2005-08-06Made the checks for category Makefiles much stricter. Only the "categoryrillig1-38/+55
normal form" passes all checks.
2005-08-06Run a quick self-test after installation.rillig1-1/+4
2005-08-06White-space fixes.rillig1-9/+9
2005-08-06Code beautification: %{$foo} instead of %$foo.rillig1-3/+3
2005-08-06Checking for C/C++ comments is not necessary for Makefiles. No-one usesrillig1-4/+1
them. All pathnames are already interpreted as being relative to the package's directory. Added a reminder that ONLY_FOR_PLATFORM should be changed to NOT_FOR_PLATFORM where appropriate.
2005-08-06Bump the PKGREVISIONs of all (638) packages that hardcode the locationsjlam1-3/+1
of Perl files to deal with the perl-5.8.7 update that moved all pkgsrc-installed Perl files into the "vendor" directories.
2005-08-06Moved the checks that had been done while loading Makefiles into arillig1-24/+29
subroutine checklines_Makefile, which is called by checkfile_package_Makefile and check_category.
2005-08-04Bump to 20050804:xtraeme2-4/+8
When using the -M flag, look in Makefile and Makefile.common files. Ok'ed by Peter Postma.
2005-08-04missed in lastabs1-1/+2
2005-08-04update pkgtools/pkg_chk to 1.60:abs2-21/+34
Make -b work without any access to pkgsrcdir
2005-08-02Update to version 0.118: pkgdiff fix and tuning.seb2-6/+11
Fix the automagic tuning of the number of unified diff context lines. It was broken since previous commit to files/pkgdiff. Also arguments '-u' and '-U NUM' to GNU diffutils 2.8.1, as seen in NetBSD 2.0_STABLE and current, are not intended to be used together. Hence only '-U NUM' is now specified. I did not reverted to the '-u -NUM' form used before as it seems undesirable. Please note that if PKGDIFF_FMT is specified in the environment, for testing or debugging purposes, we skip the automagic tuning of the number of unified diff context lines. The user probably wants to try another diff format, as suggested in the commit log of revision 1.12 of files/pkgdiff. So there is no point in adding '-U NUM' to PKGDIFF_FMT: it may just not work. See: $ diff -c -U 1 /dev/null /dev/null diff: conflicting output style options diff: Try `diff --help' for more information. While here save a few basename(1) runs.
2005-08-02Print warnings if the values of ${PKGNAME}, ${SVR4_PKGNAME},rillig1-14/+14
${DISTNAME}, ${DISTFILES}, ${EXTRACT_SUFX} cannot be resolved completely. It's better than proceeding with definitely unusable values.
2005-08-01Some more speed optimization.rillig1-10/+9