This is the log produced by jhalfs, some errors in it, but it worked well when build lfs.
-- -- http://linuxfromscratch.org/mailman/listinfo/alfs-discuss FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Le 17/02/2012 07:37, xinglp a écrit :
> This is the log produced by jhalfs, some errors in it, but it worked
> well when build lfs.
Thanks for the report. I knew that, but the reason is outside
jhalfs. It is in the Makefile of the LFS source:
-------
validxml: tmpdir
[...]
[... look at next line ...]
$(Q)rm -f appendices/*.script
$(Q)./aux-file-data.sh $(RENDERTMP)/lfs-full.xml
------
The rm line above suppresses files, which are referenced
in the book source. So when jhalfs accesses the book sources
to parse the commands scripts, those files are searched too,
and they are not found. Since they are not really needed,
xsltproc issues a warning and not an error.
As a workaround, if you have a working copy of the book
sources, you can comment out the faulty line in the
Makefile and type 'make' from the book sources directory.
Then, when you run jhalfs, use that
working copy, and it should not issue warnings.
Regards
Pierre
--
--
http://linuxfromscratch.org/mailman/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
(Continue reading)
On Fri, 17 Feb 2012 10:30:06 +0100, Pierre Labastie <pierre.labastie <at> neuf.fr> wrote: > Le 17/02/2012 07:37, xinglp a écrit : >> This is the log produced by jhalfs, some errors in it, but it worked >> well when build lfs. > Thanks for the report. I knew that, but the reason is outside > jhalfs. It is in the Makefile of the LFS source: > ------- > validxml: tmpdir > [...] > [... look at next line ...] > $(Q)rm -f appendices/*.script > $(Q)./aux-file-data.sh $(RENDERTMP)/lfs-full.xml > ------ > The rm line above suppresses files, which are referenced > in the book source. So when jhalfs accesses the book sources > to parse the commands scripts, those files are searched too, > and they are not found. Since they are not really needed, > xsltproc issues a warning and not an error. > > As a workaround, if you have a working copy of the book > sources, you can comment out the faulty line in the > Makefile and type 'make' from the book sources directory. > Then, when you run jhalfs, use that > working copy, and it should not issue warnings. I had a quick look at the Makefile and make-aux-files.sh script and couldn't see why that line is required. Bruce, is there a reason those .script files are deleted? Thanks,(Continue reading)
Matthew Burgess wrote: > I had a quick look at the Makefile and make-aux-files.sh script and couldn't > see why that line is required. Bruce, is there a reason those .script files are > deleted? It's cleanup. For jhalfs, it would be better if a different set of rules were used if different behavior is needed. -- Bruce -- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Le 17/02/2012 18:30, Bruce Dubbs a écrit : > Matthew Burgess wrote: > >> I had a quick look at the Makefile and make-aux-files.sh script and couldn't >> see why that line is required. Bruce, is there a reason those .script files are >> deleted? > It's cleanup. For jhalfs, it would be better if a different set of > rules were used if different behavior is needed. > > -- Bruce Hi, jhalfs needs an xml file to process for extracting commands scripts. 'index.xml' in book sources is the most obvious choice. But using this file leads necessary to try to find the '.script' entities, which have been deleted or do not exist. Another choice is to use lfs-full.xml, after 'make validxml'. I am currently testing that, but the target validxml depends on maketar (because aux-file-data;sh does a `ls lfs-bootscript*.bz2') and that is not specified on the 'validxml:' line in Makefile. Maybe one of you could do that (add maketar as a dependency of validxml) Thanks Pierre -- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page(Continue reading)
On Fri, Feb 17, 2012 at 4:59 PM, Pierre Labastie
<pierre.labastie <at> neuf.fr> wrote:
> Le 17/02/2012 18:30, Bruce Dubbs a écrit :
>> Matthew Burgess wrote:
>>
>>> I had a quick look at the Makefile and make-aux-files.sh script and couldn't
>>> see why that line is required. Bruce, is there a reason those .script files are
>>> deleted?
>> It's cleanup. For jhalfs, it would be better if a different set of
>> rules were used if different behavior is needed.
>>
>> -- Bruce
> Hi,
>
> jhalfs needs an xml file to process for extracting commands scripts.
> 'index.xml' in book sources is the most obvious choice. But using this file
> leads necessary to try to find the '.script' entities, which have been
> deleted
> or do not exist.
> Another choice is to use lfs-full.xml, after 'make validxml'.
> I am currently testing that, but the target validxml depends on maketar
> (because
> aux-file-data;sh does a `ls lfs-bootscript*.bz2') and that is not
> specified on the 'validxml:' line in Makefile.
> Maybe one of you could do that (add maketar as a dependency of validxml)
After looking at this all that is needed is to add the following line
to common/libs/func_book_parser right after we check out the sources.
cd ${PROGNAME}-$LFSVRS; bash process-scripts.sh >> $LOGDIR/$LOG 2>&1 ; cd ..
(Continue reading)
Thomas Pegg wrote:
> cd ${PROGNAME}-$LFSVRS; bash process-scripts.sh >> $LOGDIR/$LOG 2>&1 ; cd ..
May I suggest: pushd ${PROGNAME}-$LFSVRS; ...; popd
It's a little more robust.
-- Bruce
--
--
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
On Feb 17, 2012, at 5:22 PM, Bruce Dubbs wrote:
> Thomas Pegg wrote:
>
>> cd ${PROGNAME}-$LFSVRS; bash process-scripts.sh >> $LOGDIR/$LOG 2>&1 ; cd ..
>
> May I suggest: pushd ${PROGNAME}-$LFSVRS; ...; popd
>
> It's a little more robust.
Yes, your right, I always forget about that construct.
Thomas
--
--
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
Le 17/02/2012 23:08, Thomas Pegg a écrit :
> After looking at this all that is needed is to add the following line
> to common/libs/func_book_parser right after we check out the sources.
>
> cd ${PROGNAME}-$LFSVRS; bash process-scripts.sh>> $LOGDIR/$LOG 2>&1 ; cd ..
>
> This quells all those messages.
>
> Thomas
That is very simple, great.
I wonder if it works when using a working copy?
Maybe that command could be done just before
the xsltproc command:
pushd $BOOK > /dev/null
bash process-scripts.sh >> $LOGDIR/$LOG 2>&1
popd > /dev/null
Still I think there are the messages about packageManager.xml
--
--
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
Pierre Labastie wrote: > jhalfs needs an xml file to process for extracting commands scripts. > 'index.xml' in book sources is the most obvious choice. But using this file > leads necessary to try to find the '.script' entities, which have been > deleted or do not exist. > Another choice is to use lfs-full.xml, after 'make validxml'. > I am currently testing that, but the target validxml depends on maketar > (because aux-file-data;sh does a `ls lfs-bootscript*.bz2') and that is not > specified on the 'validxml:' line in Makefile. > Maybe one of you could do that (add maketar as a dependency of validxml) I can make a special target for you it that makes a difference, but the lfs-full.xml should do what you want. It basically just satisfies all the includes and creates one big xml file. Do you need more than that? I'm really not sure what the problem is. The current jhalfs works fine for me. -- Bruce -- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Le 17/02/2012 23:19, Bruce Dubbs a écrit : > Pierre Labastie wrote: > >> jhalfs needs an xml file to process for extracting commands scripts. >> 'index.xml' in book sources is the most obvious choice. But using this file >> leads necessary to try to find the '.script' entities, which have been >> deleted or do not exist. >> Another choice is to use lfs-full.xml, after 'make validxml'. >> I am currently testing that, but the target validxml depends on maketar >> (because aux-file-data;sh does a `ls lfs-bootscript*.bz2') and that is not >> specified on the 'validxml:' line in Makefile. >> Maybe one of you could do that (add maketar as a dependency of validxml) > I can make a special target for you it that makes a difference, but the > lfs-full.xml should do what you want. It basically just satisfies all > the includes and creates one big xml file. Do you need more than that? > > I'm really not sure what the problem is. The current jhalfs works fine > for me. Well, not really a problem with jhalfs. If somebody tries to type 'make validxml' in a just extracted LFS book source (not needed very often I agree, but this is what I tried before Thomas gave a smarter solution for jhalfs), it hangs forever: pierre <at> turboli:~/downloads/lfsbook$ make validxml Creating and cleaning /home/pierre/tmp Processing bootscripts... Validating the book... ls: cannot access lfs-bootscripts*.bz2: No such file or directory [hangs](Continue reading)
On Fri, 2012-02-17 at 23:55 +0100, Pierre Labastie wrote: > pierre <at> turboli:~/downloads/lfsbook$ make validxml > Creating and cleaning /home/pierre/tmp > Processing bootscripts... > Validating the book... > ls: cannot access lfs-bootscripts*.bz2: No such file or directory > [hangs] > > The reason is that the line: > > validxml: tmpdir (in Makefile) > > should be: > > validxml: maketar tmpdir Yeah, it's the call out to md5sum in aux-file-date.sh that actually hangs because $bootscript is the null string because the `ls` failed as you quoted above. I'm not sure your fix is right though. Being a pedant, the last 2 commands in the validxml target have nothing to do with validating the XML; that is obviously achieved by the call to xmllint. Then again, those last 2 commands are required in order to *produce* valid XML for other targets to then make use of. I think the proper fix is to remove the maketar target completely, and move the call to make-aux-files.sh into the aux-file-data.sh script. My reasoning is that you can't even call ./aux-file-data.sh from the(Continue reading)
Le 18/02/2012 00:49, Matt Burgess a écrit : > > I'm not sure your fix is right though. Being a pedant, the last 2 > commands in the validxml target have nothing to do with validating the > XML; that is obviously achieved by the call to xmllint. Then again, > those last 2 commands are required in order to *produce* valid XML for > other targets to then make use of. Actually I suggested the simplest thing I could think of. I agree that other solutions are cleaner. > > As an aside, MAKETAR is never defined anywhere, so the guard doesn't > appear to ever do anything. > > Well, MAKETAR could be defined on the command line, for example: make MAKETAR=n Don't know if it is of some use though... And it is not documented anywhere. Regards, Pierre -- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page(Continue reading)
On Fri, 2012-02-17 at 23:49 +0000, Matt Burgess wrote: > Suggested patch attached. Applied in r9762. Regards, Matt. -- -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
Pierre Labastie wrote: > Le 17/02/2012 07:37, xinglp a écrit : >> This is the log produced by jhalfs, some errors in it, but it worked >> well when build lfs. > Thanks for the report. I knew that, but the reason is outside > jhalfs. It is in the Makefile of the LFS source: > ------- > validxml: tmpdir > [...] > [... look at next line ...] > $(Q)rm -f appendices/*.script > $(Q)./aux-file-data.sh $(RENDERTMP)/lfs-full.xml > ------ > The rm line above suppresses files, which are referenced > in the book source. So when jhalfs accesses the book sources > to parse the commands scripts, those files are searched too, > and they are not found. Since they are not really needed, > xsltproc issues a warning and not an error. > > As a workaround, if you have a working copy of the book > sources, you can comment out the faulty line in the > Makefile and type 'make' from the book sources directory. > Then, when you run jhalfs, use that > working copy, and it should not issue warnings. It is not a faulty line. It forces the scripts to be put into a form usable by the xml. If there is a change, they won't get incorporated into the book without that line. -- Bruce(Continue reading)
xinglp wrote:
> This is the log produced by jhalfs, some errors in it, but it worked
> well when build lfs.
>
/home/≤somepath>/lfs_xml/appendices/scripts.xml:58: parser error :
Entity 'rc' not defined
<screen>&rc;</screen>
Actually the entity rc is defined in the scripts.xml file, line 16.
/home/≤somepath>/lfs_xml/chapter07/site.xml:33: parser error : Entity
'site' not defined
<screen role="auto">&site;</screen>
Entity site is defined in line 5 of site.xml.
You are somehow not picking up the entity definitions. However, in all
cases, they appear insite <screen>...</screen> tags. You could get
around most of these by just skipping ove the appendices.
-- Bruce
--
--
http://linuxfromscratch.org/mailman/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
RSS Feed