load the base set of files first, then any user arch/dist specific override
files second.
this is important since whenever config items get moved in the base files,
this can break the override files, if an option specified in one was moved
to a different common file that it loaded after.
Gbp-Dch: Ignore
these should be bypassing the possibility of running the auto script
surely, just like how this is done for `--all`, otherwise you pass through
the auto file twice.
Gbp-Dch: Ignore
- detect lack of options using actual arg count rather than first arg
being an empty string.
- fix string splitting issues by looping properly on "${@}"
- tidier
Gbp-Dch: Ignore
...alongside printing usage (which is perhaps unnecessary), so that it is
actually clear to users that a problem occurred, and what.
and capture it before option processing of remaining args.
Gbp-Dch: Ignore
- it's only used by the debootstrap script after alternatives were dropped
long ago, so let's move it, avoiding it being loaded for everything
else.
- there's no need to pass printing another message through it.
- there's little point in making the sid distinction if you happen to
decide to build sid, it's a given that it's less stable than stable.
really, is there any need for this at all?
Gbp-Dch: Ignore
the default grub config has a quirk of using the syslinux splash when
present, which as a sidenote i believe is because only the syslinux script
has the code for converting an SVG (i've not researched to confirm).
the problem is that the default splash is such, with the text listing
package versions used, that this overlaps with the menu title in grub.
i've tweaked the "built on" and package details text to reduce its size
(bumped the line-height up slightly though) to avoid the overlap.
this was a pain in the *** to get done. inkscape likes to crash at the
slightest little thing like clicking on interface controls; each time it
saves it seems to swap a bunch of stuff in the file which can make a
hugely messy diff, i had to save, close, open, change, undo, and save
again, just to get a sain diff; and numerous times testing different sizes
and such in actual builds (rather than separate conversions) some of the
package detail lines would for some unknown reason decide to end up on the
same line as another, wth...
anyway, i've finally managed to get a new revision successfully created
and it seems bug free wrt. that same line issue to me, both converting
outside of live-build (in my sid host), and within a buster build. the size
of the text changed seems closer to the size of the actual menu entries
(there's no need for it to be bigger), and now is nicely clear of the menu
title.
Gbp-Dch: Short
opening the file in inkscape resulted in a dialog explaining that inkscape
used to work on a 90 DPI basis but switched to 96 DPI for better
compatibility reasons, and wanted to convert the file.
this is simply the result of it having done the conversion.
it advised that for documents not intended to be printed that there should
be no noticeable difference. i checked the output and it seems identical
to before. i tried to look at the actual diff, but it's a lot of noise,
most of which is just reordering, and i gave up, since the output looks
identical i'm not going to worry.
Gbp-Dch: Ignore
in tweaking a previous commit to remove some excessive change before
submission, i mistakenly identified the part of the sed replacement
restored here as being unnecessary to its functionality, but in fact it is.
without it the placeholder is not actually removed.
the lack of removal of the placeholder meant that you ended up with
duplicate copies of the live menu entries.
Gbp-Dch: Ignore
more fitting now that we've moved the advanced installer entries
out to a different submenu, leaving just memtest (and HDT on
syslinux).
the advanced.cfg file is also renamed to utilities.cfg in the syslinux
case, but in a backwards compatible way of moving the user advanced.cfg
file over the new one, if the user provides a file with the old name.
alternatively we could just leave the old name in place, but that would be
a little odd.
Gbp-Dch: Short
- most entries used vga=788 rather than vga=normal (which are equivalent)
- the syslinux menu uses 788 exclusively
- so do the official install discs
Gbp-Dch: Short
This is considered to be more robust.
Two instances remain:
scripts/build/chroot_archives, line 257:
if [ "${LB_APT}" = "aptitude" ] && [ ! $(Chroot chroot "which aptitude") ]
The command is run inside a chroot where the environment might be special,
and would need further testing.
manpages/Makefile, line 42:
@if [ ! -x "$$(which po4a 2>/dev/null)" ]; \
I am insufficiently familiar with makefile syntax to edit this.
instead of cramming the live entries into a string via a layer of functions
for terminating entries with newlines, which we then have to run through
perl to tweak the newlines for correct use with sed... let's write the
entries to a temp file, then use that file in the sed replacement.
the helper functions injecting newlines to the end of entries as they were
built into a long string have obviously become unnecessary and so were
removed. one function was renamed for reasons of consistency and clarity.
the file is initially deleted before use for reasons of wanting to bullet
proof the codebase to work properly under conditions of recovering from
failure/cancellation, `--force` re-running and such.
this removes the last use of perl.
Gbp-Dch: Short
support for this has not existed since v2.0~a24-1 !!
until 4d22ca948a (v2.0~a24-1), `PROGRAM` was
defined as `PROGRAM="$(basename ${0})"`. then this commit simply got rid
of the definition.
it then flipped back and forth a bit in a few ways, but always sticking
to a simple fixed "live-build" type string after that commit, as it has
been ever since.
the code removed here has thus been junk since then, and the documentation
corrected here out of date since then.
the checking of `LB_CONFIG` is also possibly junk, but i know too little
about it. it was introduced in v1.0~a43-1 in commit
705a4178e7 with a changelog entry of
"Centraly evaluate LH_CONFIG for custom configuration file."
Gbp-Dch: Short
it reported: "possible bashism in scripts/build/binary_loopback_cfg line 284 (should be '.', not 'source')"
which is clearly a misidentification.
Gbp-Dch: Ignore
also enables greater flexibility for users to override more aspects of
theme separate from menu entries, and similarly the initial config details
separate from the menu entries. (with the recent change to merge user files
with the defaults rather than replace them, users can pick which of these
individual components they wish to customise.
identical to change as sent in to Kali and accepted by Raphaël.
Gbp-Dch: Short
partly lost in some adjustments that were made to the submitted work,
which was focused on restoring the 'start installer' entry.
there is no need for dynamic setting of these two `source` imports in the
default file, in fact user modifications should also use the fixed import
commands in future.
note that the old placeholders however remain replaced, which inject
precisely this string, for backwards compatibility.
Gbp-Dch: Ignore
during testing i encountered an unexpected error resulting from the
following condition:
- bootstrap was cached
- cache of bootstrap packages was empty (from playing with `lb clean`)
- installer was not none|live
everything works fine up until the main installer script, which comes to
an abrupt halt with an error due to the missing cached bootstrap packages
that it wants to copy.
this situation is easy to unintentionally create, as i managed to do.
here we catch the failure condition, correcting for it.
this is done by checking for the missing cached packages when restoring
the bootstrap from the cache, and skipping this if the packages were
missing, thus forcing the bootstrap to be rebuilt. the packages should
then be found within the cache, allowing the installer stage to complete
successfully.
of course the bootstrap stage will only cache the packages if caching is
enabled, but if caching is disabled and installer enabled, the config
validation will catch that, reporting that problem.
Gbp-Dch: Short
install mode was silent when selinux was not enabled, whilst remove mode
always output a "Begin unmounting..." message. this makes both modes silent
when selinux is not enabled.
this also happens to fix an unintended side effect of
d79c96232b whereby a warning subsequently
was always emitted in remove mode when selinux was not in use, which was
not ideal.
Gbp-Dch: Short
not worth listing the specifics
mostly the following though:
- "(comma or space separated)" -> "(a comma or space separated list)"
- quoting values
Gbp-Dch: Ignore
The unconditional SVG to PNG conversion could overwrite a splash.png
provided by the user. Ensure we don't overwrite such a file. But we
still remove the SVG file as syslinux is not able to make use of it.
Lyndon Brown
johnraff
Raphaël Hertzog