On Sat, 2020-04-04 at 20:58 -0500, Bruce Dubbs via blfs-dev wrote: > On 4/4/20 8:33 PM, Ken Moffat via blfs-dev wrote: > > On Sat, Apr 04, 2020 at 07:41:45PM +0100, Ken Moffat via blfs-dev > > wrote: > > > On Sat, Apr 04, 2020 at 12:54:57PM -0500, Bruce Dubbs via blfs- > > > dev wrote: > > > > On 4/4/20 12:15 PM, Ken Moffat via blfs-book wrote: > > > > > [ Cc: -dev added ] > > > > > > > > > > I left it open because I hope there is some better way of > > > > > keepign > > > > > the version, tarball size and md5sum in-sync. I sort of > > > > > assumed > > > > > that using e.g. md5sum from one package in another would > > > > > break > > > > > things (don't have time to check that at the moment). > > > > > > > > Well the md5sum and package size won't change. We kinda to the > > > > same thing > > > > with bind and bind-utilities. They are two different builds of > > > > the same > > > > tarball. > > > > > > > > In a similar situation we have qt5 and QtWebEngine. Although > > > > 'webengine' is > > > > available as a separate tarball, we almost always build both at > > > > each new > > > > release. > > > > > > > > The problem with js is if some package, e.g. seamonkey or > > > > thunderbird, does > > > > not keep up with firefox. We also have the issue of polkit > > > > still at js60. > > > > I checked into that and polkit does not build with js68. > > > > However it does > > > > seem that there is only one polkit file that uses js. > > > > > > > > > > Thanks for the pointer to bind and bind-utils : I'll change JS68 > > > and > > > packages. > > > > > And now I'll note tht it isn't an ideal solution - copying > > firefox-size into js68 worked, but only because it was (re) defined > > in js68. If I comment that part, js68 does not validate because > > firefox-size is not defined (i.e. the definition is local to each > > xml file). > > > > What I was hoping for was some way of defining it once, so that > > both > > users can only show the same value. I suppose putting the size and > > md5sum in packages could do that (ditto for bind), but that would > > look messy. > > > > I'm also grumpy that I've now got to measure js68 when I update > > firefox, but at least js68 can be used generally after the next > > polkit release happens. > > OK, Grumpy. We could try to use common md5sums by creating entities > in > general.ent, but that would make things inconsistent. Since jsxx > needs > to be rebuilt anyway for timing and build size, is it really that > hard > to just copy the md5sum into the page? I double check the md5sums > for > all new packages anyway, so a mistake can be caught quickly. >
What can be done is having small .ent files for entities common to several .xml sources, like ff-js-common.ent containing: <!ENTITY ffjs-download-http "..."> <!ENTITY ffjs-download-ftp " "> <!ENTITY ffjs-md5 "..."> <!ENTITY ffjs-size "..."> then in say, firefox.xml: <!DOCTYPE sect1 PUBLIC "-//OASIS//DTD DocBook XML V4.5//EN" "http://www.oasis-open.org/docbook/xml/4.5/docbookx.dtd"; [ <!ENTITY % general-entities SYSTEM "../../general.ent"> %general-entities; <!ENTITY % ffjs-entities SYSTEM "../../ff-js-common.ent"> %ffjs-entities; <!ENTITY firefox-buildsize "..."> <!-- etc --> That's how XML works normally. True, it's verbose :) Pierre -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
