make-w32
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Latest CVS, Increased win32 test failures since beta -> 33


From: J. Grant
Subject: Re: Latest CVS, Increased win32 test failures since beta -> 33
Date: Sat, 30 Jul 2005 12:07:54 +0100
User-agent: Mozilla/5.0 (X11; U; GNU/Linux i686; en; rv:1.7.8) Gecko/20050513 Debian/1.7.8-1

Hi Paul,

[...]
  jg> misc/general3 ........................................... FAILED (9/10 
passed)

This failure is almost certainly due to the change in backslash-newline
handling, since misc/general3 is where I added all my testing.

Ok, I have attatched the files from general3, test 4 which failed

  jg> Tests which only fail because of incorrectly created base output file:

  jg> features/recursion
  jg> options/dash-I.mk
  jg> targets/DEFAULT
  jg> variables/CURDIR
  jg> variables/MAKE

In what way were they incorrectly created?

These wwere either the "echo" bug ("/c:/" instead of "c:/" generated),

or the:

"c:/msys/1.0/home/" vs. "/home/" style paths at the begining of the output.


See example echo_var_MSYS_bug.mak  attatched. (output included)


broken_MSYS_echo_path.mak is also attatched, this is another problem I noticed.

  jg> These are the tests which I think we should try and find solutions for:

  jg> work/features/export.mk UNEXPORT_ALL=1
  jg> features/export.mk EXPORT_ALL=1 UNEXPORT_ALL=1
  jg> features/patspecific_vars.mk
  jg> functions/abspath
  jg> functions/value
  jg> functions/wildcard.mk print2
  jg> functions/wildcard.mk clean
  jg> variables/special

Don't know about these.

Hmm, what don't you know about them specifically..? Why they are failing
etc? I will email you the 41KB zip of my make/tests/work directory. (If anyone else would like this zip please email me)

  jg> options/dash-l.mk -l 0.0001 -j 4

I don't think Windows supports the -l option so I doubt that this test
should even be invoked.

Ok, I don't know where the code is to disable tests, would you be able
to add this to the disabled list please...

I've make some changes (checked in just tonight) that tries to handle
environment variables more safely than previous versions, just in case
some variables are leaking from test to test (I don't know if Windows
allows that to happen).

Could you disable the symlink tests as well please?  When these changes
are all in I will update my code and run the tests again.

Kind regards
JG
foo\
bar
foo\
bar
foo\
    bar
foo\
    bar
foo \
bar
foo \
bar
foo \
    bar
foo \
    bar
*** work/misc/general3.base.4   Sat Jul  9 00:47:30 2005
--- work/misc/general3.log.4    Sat Jul  9 00:47:30 2005
***************
*** 1,16 ****
! foo\
! bar
! foo\
! bar
! foo\
!     bar
! foo\
!     bar
! foo \
! bar
! foo \
! bar
! foo \
!     bar
! foo \
!     bar
--- 1,14 ----
! foo\ bar
! foo\ bar
! foo\
!     bar
! foo\
!     bar
! foo \
! bar
! foo \
! bar
! foo \
!     bar
! foo \
!     bar
foo\ bar
foo\ bar
foo\
    bar
foo\
    bar
foo \
bar
foo \
bar
foo \
    bar
foo \
    bar
all:
        @echo 'foo\
bar'
        @echo 'foo\
        bar'
        @echo 'foo\
    bar'
        @echo 'foo\
            bar'
        @echo 'foo \
bar'
        @echo 'foo \
        bar'
        @echo 'foo \
    bar'
        @echo 'foo \
            bar'
all:
        @echo 'maketemp="$(MAKE)"'
        echo $(MAKE)

# output is as follows: (see broken maketemp with \)

# $ make -f echo_var_MSYS_bug.mak 
#maketemp=\C:/msys/1.0/local/bin/make"
#echo C:/msys/1.0/local/bin/make
#C:/msys/1.0/local/bin/make

all:
        /bin/echo hello

#output looks like this
# /bin/echo hello
#process_begin: CreateProcess((null), /bin/echo hello, ...) failed.
#make (e=3): The system cannot find the path specified.
#make: *** [fubar] Error 3

reply via email to

[Prev in Thread] Current Thread [Next in Thread]