Opened 5 years ago

Closed 5 years ago

Last modified 4 years ago

#4607 closed defect (fixed)

Makefile does not generate a correct nochuncks book

Reported by: Pierre Labastie Owned by: lfs-book
Priority: normal Milestone: 10.0
Component: Book Version: SVN
Severity: normal Keywords:
Cc:

Description

Reported by K. Buckley on -dev:

the noChunks book doesn't
"see" any changes made by aux-file-data.sh and so ends up
rendering, for example, in Chapter 3's package listing

  LFS-Bootscripts (20191031) - BOOTSCRIPTS-SIZE KB:

rather than

  LFS-Bootscripts (20191031) - 32 KB:

The fix is very simple, remove the "2" of lfs-html2.xml in the instructions:

        $(Q)xsltproc --nonet                                \
                --stringparam rootid "$(ROOT_ID)"      \
                --output $(BASEDIR)/$(NOCHUNKS_OUTPUT) \
                stylesheets/lfs-nochunks.xsl           \
                $(RENDERTMP)/lfs-html2.xml

But Bruce said this would have some impact on the scripts for rendering the book, so this needs some clarification.

Change History (9)

comment:1 by Bruce Dubbs, 5 years ago

Let's see if we can understand this.

Both book: and nochunks have identical validate and profile-html perrequsites.

validate: outputs $(RENDERTMP)/lfs-html2.xml and $(RENDERTMP)/lfs-full.xml; it then runs aux-file-data.sh against $(RENDERTMP)/lfs-full.xml

profile-html: uses $(RENDERTMP)/lfs-full.xml and outputs $(RENDERTMP)/lfs-html.xml

book: uses $(RENDERTMP)/lfs-html.xml to render the html files.

nochunks: uses $(RENDERTMP)/lfs-html2.xml to render the html files


I committed revision 11773. See if that works. (s/html2/full/ in nochunks)

comment:2 by Pierre Labastie, 5 years ago

I was wondering why you used lfs-full instead of lfs-html, so I did a diff of the generated books with one and the other. The only changes are automatically generated ID refs, so not visible.

Now I wonder why we have the profile-html target at all, since it does not seem to change the output... Doing a diff of lfs-full.xml and lfs-html.xml, I see two main changes (there may be more, the output is big):

  • the sect1info nodes are removed
  • the character entities (such as &#A9;) are replaced by the characters themselves (© in the example)

So maybe it is useful for some enconding combinations, but if it is useful for chunked output, why not use it for non chunked?

Version 0, edited 5 years ago by Pierre Labastie (next)

comment:3 by Bruce Dubbs, 5 years ago

I used lfs-html.xml because that is what book: uses.

I really don't want to make any other changes. What we have works, even if it is not perfect.

in reply to:  3 comment:4 by Pierre Labastie, 5 years ago

Replying to bdubbs:

I used lfs-html.xml because that is what book: uses.

I really don't want to make any other changes. What we have works, even if it is not perfect.

I do not want to make other changes either, but you used lfs-full.xml, not lfs-html.xml. So I wondered why... lfs-html.xml was my first proposition, actually.

comment:5 by Bruce Dubbs, 5 years ago

Ow! My mistake. I did want to change to lfs-html.xml. Updated at revision 11774.

comment:6 by Pierre Labastie, 5 years ago

Wrong place (you modified the pdf: target). Let me fix it.

comment:7 by Pierre Labastie, 5 years ago

Resolution: fixed
Status: newclosed

Eventually fixed at r11775

comment:8 by Bruce Dubbs, 5 years ago

Thanks. It's an unusual day for me. Perhaps I ought to take the day off...

comment:9 by Bruce Dubbs, 4 years ago

Milestone: 9.210.0

Milestone renamed

Note: See TracTickets for help on using tickets.