zeromq / gsl Goto Github PK

Thought I would mention this, you just need to run

$ brew install pcre

And follow the rest of the instructions. Just saying, because darwin is not unix...

Hi,

I ran into really weird issue with gsl, zproject and travis target. So far the minimal reproducer case is

    <project
        name = "dbg"
        script = "zproject.gsl"
        repository = "https://github.com">

    <target name = "redhat" />
    <target name = "travis" />
    
   &lt;/project>

Running gsl file.xml works well. But once I change repository content, like "https://bitbucket.org", "https://gitlab.com", "[email protected]", "https://github.co.uk", then you will get gsl error

gsl/4 M: Generating skeleton .travis.yml script

Memory tag error - 0x563c29d2f640 - <Unknown> (0)
Aborted (core dumped)

Used dependencies (I have OBS packages installed from git-draft project)

rpm -q zproject generator-scripting-language

zproject-1504557118.26bc68b-77.1.x86_64

generator-scripting-language-1504755625.c020d42-13.1.x86_64

... a small piece of work for taking a community with you.

Moving to my new mac (MacOS Monterrey 12.1) and trying to install getting 'pcre.h' file not found error. I have installed and then reinstalled pcre, also updated Xcode's command line tools but still having no love. Full error text below.

ggpcre.c
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/c++/4.2.1
ggpcre.c:54:10: fatal error: 'pcre.h' file not found
#include <pcre.h>
^~~~~~~~

I fixed the "not-escaping the escape symbol" in the README.md by naïvely modifying the file instead of building with gitdown. I am not sure if that is important, but thought I would file a new issue.

Invoke method thread . send() with argument leads to an error.
Can you help me?

File: main.gsl

hello_script = file.slurp("./hello.gsl", error)
if defined(error)
    echo error
endif

hello_thread = thread.new(hello_script, error)
if defined(error)
    echo error
else
    hello_thread.send("message")
endif

File: hello.gsl

thread.receive(parent, message)
echo message

Run:

gsl main.gsl

Got the error: Parameter mismatch for function: hello_thread.send

My IDE insist of saving XML files as UTF8 with BOM, as I don't use any special characters it is actually a regular ascii which gsl can process.

However GSL cannot process the BOM at the beginning of the file.

Any chance supporting UTF8 with BOM? If someone can direct me to the correct place I can try and solve it myself.

Thanks.

CCNAME='arm-linux-gnueabihf-gcc' ./c -q -r libgsl.a ggobjt.o ggcomm.o ggscop.o ggcode.o ggcodem.o ggpars.o ggparsm.o ggconv.o ggdiag.o ggenvt.o ggfile.o gggsl.o ggmath.o ggpcre.o ggproc.o ggscrp.o ggsock.o ggstrn.o ggsymb.o ggthrd.o ggtime.o ggxml.o sflbits.o sflcvds.o sflcvts.o sflfile.o sfllbuf.o sflnode.o sflsock.o sfltron.o smtlog.o smtmsg.o smtsimu.o smttst2.o sflcomp.o sflcvns.o sfldate.o sflfind.o sfllist.o sflprint.o
sflsort.o sfluid.o smtoper.o smtslot.o smttst3.o sflcons.o sflcvsb.o sfldes.o sflfort.o sflmail.o sflproc.o sflstr.o sflxml.o smtpipe.o smtsmtp.o smtxlog.o sflconv.o sflcvsd.o sfldescr.o sflheap.o sflmath.o sflscrit.o sflsymb.o sflxmll.o smtpop.o smtsock.o sflcryp.o sflcvsn.o sfldir.o
sflhttp.o sflmem.o sflsearch.o sflsyst.o sflxmls.o smtrdns.o smttime.o sflcvbs.o sflcvst.o sflenv.o sflini.o sflmesg.o sflsha.o sfltok.o smtecho.o smtrdnsl.o smttran.o sflcvdp.o sflcvtp.o sflexdr.o sfllang.o sflmime.o sflslot.o sfltree.o smtlib.o smtserv.o smttst1.o
CCNAME='arm-linux-gnueabihf-gcc' ./c -q -specs=/void-packages/common/environment/configure/gccspecs/hardened-cc1  -fstack-protector-strong -D_FORTIFY_SOURCE=2  -O2 -pipe -march=armv6 -mfpu=vfp -mfloat-abi=hard   -I/usr/arm-linux-gnueabihf/usr/include -specs=/void-packages/common/environment/configure/gccspecs/hardened-ld -Wl,-z,relro -Wl,-z,now -Wl,--as-needed     -L/usr/arm-linux-gnueabihf/usr/lib  -L gsl
./gsl -v
./gsl: ./gsl: cannot execute binary file
make: *** [Makefile:188: everything] Error 126

Even though the build is successful, being unable to execute across architectures makes gsl impractical to cross-build.

I posted this already in the imatrix gsl repo but got no answer. It is basically a first crude patch for CDATA handling. Here the original post:
(imatix#98)

I have the same problem and written up a crude patch as an example how it could be implemented.

If such an addition would be welcome I could then modify the patch in order to add more checks make it more robust etc.. If yes, I would appreciate some comments on how the same functionality can be archived with more functionality that is already available in gsl.

https://gist.github.com/MaxSagebaum/2c2e3443ca36e093e65025692bfc1220#file-0001-hanlding-of-cdata-as-one-single-value-item-patch

If GSL was 25 years ahead of its time Rational Rose was a century ahead of its time.
Meta & data as a dialect of LISP, with a graphical front end, and the source code generation.
GSL is to be commended indeed, but XML legacy is a dead weight around its neck. The usability of the whole GSL solution is on the level of an academic research project. GSL needs a push to make it into a product. Just like RR was. (Un)fortunately 100 years ahead of its time...

This is due to the fuzzy matching on gcc, and previously due to "$CCNAME" = "gcc"

The fix I used prior was the following patch:

--- src/c.fix   2017-02-10 21:50:13.882583271 -0500
+++ src/c       2017-02-10 21:55:55.064681445 -0500
@@ -154,7 +154,7 @@
 #
 #  Generic modern GCC system
 #
-if [ "$CCNAME" = "gcc" ]; then
+if [ "$(${CCNAME} --version | grep -c \(GCC\))" -gt 0 ]; then
     [ -z "$BOOM_MODEL_NOOPT" ] && CCDEBUG="-O2"
     [ -z "$BOOM_MODEL_NOOPT" ] && CCNODEBUG="$CCNODEBUG -O2"
     CCOPTS="-D_REENTRANT -D_GNU_SOURCE -Wall -Wno-unused -fno-strict-aliasing"

This is somewhat heavier, but this also works on OpenBSD where "${CNAME}" = "cc".

I intend to re-work this patch to be able to continue to package gsl-ucg for my distribution.

Hi there,
I struggle with passing attributes via cli to a script. In the documentation it says:

gsl -script:interest deposits.xml

All variables that we define and use are stored in the data tree, somewhere.

'Somewhere' is a bit unspecific ;) but I guess it is stored in the datatree toplevel-element as you would also override the script-attr that way...

So giving this very basic script (test.gsl):

.template 0
echo "attribute: $(myattribute?'NO ATTRIBUTE SET')"
.endtemplate

and calling it via

gsl -myattribute:TEST test.gsl

Results in NO ATTRIBUTE SET.

Strange thing, if calling via xml-file that have a script-attribute, you actually can override the script being used via -script:otherscript but if echoing out the value of $(script), the value is still the unaltered one....

Any hints?

From the pcre/NEWS file is looks like it's version 6.5 from 01-Feb-06, more than 10 years ago. PCRE has had a few CVEs in the meanwhile.

Thanks to @jimklimov work now at least on Linux we can build and link against the system library rather than the embedded one, but for Windows and other OSes users this problem should really be fixed.

I would like to use regexp.match(pattern, text, match) in order to manipulate the text string.
For example, in case that text = "completeProfile", I want to obtain "COMPLETE_PROFILE" as result.

pattern is just a regular expression, I can figure it out, but my question is: where is stored the resulted string? I can't figure out how to access the resulted string (in my case "COMPLETE_PROFILE") in my script.

Or I am trying to use a wrong function? regexp.match() is not what I should use for what I want to achieve? I am very curious about it, please let me know.