That's the response to:
tdecmshell fonts
This is far from the first time I've had to ask this, but I can never remember how to locate anything that provided an answer, likely because I've asked on IRC, where I have no archive, instead of here. Other than kinfocenter5, I don't see any package or menu item containing infocenter either. It's not missing from the KControl window. tdecmshell xserver does work.
I have another installation, apparently with all the same trinity packages installed, which is running tdecmshell fonts. lsof | grep cmshell shows only:
/opt/trinity/bin/tdecmshell /opt/trinity/lib64/libtdeinit_tdescmshell.so
These exist on both systems.
What package contains the fonts module in 14.0.x? Why does it run on one but not the other?
--- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 7:21 AM, Felix Miata mrmazda@earthlink.net wrote:
That's the response to:
tdecmshell fontsThis is far from the first time I've had to ask this, but I can never remember how to locate anything that provided an answer, likely because I've asked on IRC, where I have no archive, instead of here. Other than kinfocenter5, I don't see any package or menu item containing infocenter either. It's not missing from the KControl window. tdecmshell xserver does work.
I have another installation, apparently with all the same trinity packages installed, which is running tdecmshell fonts. lsof | grep cmshell shows only:
/opt/trinity/bin/tdecmshell /opt/trinity/lib64/libtdeinit_tdescmshell.so
These exist on both systems.
What package contains the fonts module in 14.0.x? Why does it run on one but not the other?
if you have one working system and one not, install strace and run this on both systems:
$ strace -o log.txt -ff tdecmshell fonts
you can then do a "diff" on the two sets of log.txt.NNNNN files and it should be blindingly obvious where one file significantly differs from the other. it's going to be awkward because the files will be pretty different: you could instead put the two files side-by-side in xterms of identical height and do page-down page-down page-down instead.
l.
deloptes composed on 2018-06-03 09:26 (UTC+0200):
Luke Kenneth Casson Leighton wrote:
you could instead put the two files side-by-side in xterms of identical height and do page-down page-down page-down instead.
there is kdiff3 for that
It does a nice job of showing off differences. I see lots of differences related to which fonts are or are not installed, and the different hostnames and various addresses, but I fail to ID any missing bit to ID why 760's 'tdecmshell fonts' fails and 780's succeeds.
http://fm.no-ip.com/Tmp/Linux/KDE/stracelogs.zip
On Sun, Jun 3, 2018 at 9:27 AM, Felix Miata mrmazda@earthlink.net wrote:
deloptes composed on 2018-06-03 09:26 (UTC+0200):
Luke Kenneth Casson Leighton wrote:
you could instead put the two files side-by-side in xterms of identical height and do page-down page-down page-down instead.
there is kdiff3 for that
It does a nice job of showing off differences. I see lots of differences related to which fonts are or are not installed, and the different hostnames and various addresses, but I fail to ID any missing bit to ID why 760's 'tdecmshell fonts' fails and 780's succeeds.
you'll be looking for the first "major" obvious difference. "the font is missing" is f****g useless: you already know that. it will be something like a directory name, or a .so missing, or a .so loaded from a *different directory* or... a config file that is missing or... basically something right at the start.
by the time you get to "the fonts are not loaded"... well... you already know that, so you're clearly *not* looking for that...
... *UNLESS*....
the locations where each system is *looking* for the fonts is different. *that* is something you will want to know.
track and *interpret* the differences... don't just go "errr there are differences". work backwards for what the *important* differences are.
l.
open("/usr/share/locale/locale.alias", O_RDONLY|O_CLOEXEC) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=2939, ...}) = 0 -read(3, "# Locale name alias data base.\n#"..., 4096) = 2939 -read(3, "", 4096) = 0 +read(3, "# Locale name alias data base.\n#"..., 1024) = 1024 +read(3, "nd for the time being for\n# back"..., 1024) = 1024 +read(3, "59-1\ngalego\t\tgl_ES.ISO-8859-1\nga"..., 1024) = 891 +read(3, "", 1024) = 0 close(3) = 0
first major difference: locale not set.
--- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 10:56 AM, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
On Sun, Jun 3, 2018 at 9:27 AM, Felix Miata mrmazda@earthlink.net wrote:
deloptes composed on 2018-06-03 09:26 (UTC+0200):
Luke Kenneth Casson Leighton wrote:
you could instead put the two files side-by-side in xterms of identical height and do page-down page-down page-down instead.
there is kdiff3 for that
It does a nice job of showing off differences. I see lots of differences related to which fonts are or are not installed, and the different hostnames and various addresses, but I fail to ID any missing bit to ID why 760's 'tdecmshell fonts' fails and 780's succeeds.
you'll be looking for the first "major" obvious difference. "the font is missing" is f****g useless: you already know that. it will be something like a directory name, or a .so missing, or a .so loaded from a *different directory* or... a config file that is missing or... basically something right at the start.
by the time you get to "the fonts are not loaded"... well... you already know that, so you're clearly *not* looking for that...
... *UNLESS*....
the locations where each system is *looking* for the fonts is different. *that* is something you will want to know.
track and *interpret* the differences... don't just go "errr there are differences". work backwards for what the *important* differences are.
l.
read(5, "<?xml version='1.0'?>\n<!DOCTYPE "..., 8192) = 3364 read(5, "", 8192) = 0 close(5) = 0 -getrandom("\xd8\xfc\x79\x69\xc3\xed\x3f\x71", 8, GRND_NONBLOCK) = 8 -stat("/root/.config/fontconfig/rendering-options.conf", {st_mode=S_IFREG|0664, st_size=517, ...}) = 0 -stat("/root/.config/fontconfig/rendering-options.conf", {st_mode=S_IFREG|0664, st_size=517, ...}) = 0 -access("/root/.config/fontconfig/rendering-options.conf", R_OK) = 0 -stat("/root/.config/fontconfig/rendering-options.conf", {st_mode=S_IFREG|0664, st_size=517, ...}) = 0 -openat(AT_FDCWD, "/root/.config/fontconfig/rendering-options.conf", O_RDONLY|O_CLOEXEC) = 5 -read(5, "<?xml version=\"1.0\"?>\n<!DOCTYPE "..., 8192) = 517 -read(5, "", 8192) = 0 -close(5) = 0 -getrandom("\x20\xa3\xc2\x43\x63\x24\x77\xb2", 8, GRND_NONBLOCK) = 8 -access("/root/.config/fontconfig/rendering-options.conf", R_OK) = 0 +getrandom("\xb3\x97\xce\xa5\xef\x3f\xce\xd0", 8, GRND_NONBLOCK) = 8 +stat("/root/.config/fontconfig/rendering-options.conf", 0x7fff190f8e30) = -1 ENOENT (No such file or directory) +stat("/root/.config/fontconfig/rendering-options.conf", 0x7fff190f8e30) = -1 ENOENT (No such file or directory) +access("/root/.config/fontconfig/rendering-options.conf", R_OK) = -1 ENOENT (No such file or directory) +access("/root/.config/fontconfig/rendering-options.conf", R_OK) = -1 ENOENT (No such file or directory)
second major difference: /root/.config/fontconfig/rendering-options.conf is not empty on one of the systems.
--- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 11:00 AM, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
open("/usr/share/locale/locale.alias", O_RDONLY|O_CLOEXEC) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=2939, ...}) = 0 -read(3, "# Locale name alias data base.\n#"..., 4096) = 2939 -read(3, "", 4096) = 0 +read(3, "# Locale name alias data base.\n#"..., 1024) = 1024 +read(3, "nd for the time being for\n# back"..., 1024) = 1024 +read(3, "59-1\ngalego\t\tgl_ES.ISO-8859-1\nga"..., 1024) = 891 +read(3, "", 1024) = 0 close(3) = 0
first major difference: locale not set.
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 10:56 AM, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
On Sun, Jun 3, 2018 at 9:27 AM, Felix Miata mrmazda@earthlink.net wrote:
deloptes composed on 2018-06-03 09:26 (UTC+0200):
Luke Kenneth Casson Leighton wrote:
you could instead put the two files side-by-side in xterms of identical height and do page-down page-down page-down instead.
there is kdiff3 for that
It does a nice job of showing off differences. I see lots of differences related to which fonts are or are not installed, and the different hostnames and various addresses, but I fail to ID any missing bit to ID why 760's 'tdecmshell fonts' fails and 780's succeeds.
you'll be looking for the first "major" obvious difference. "the font is missing" is f****g useless: you already know that. it will be something like a directory name, or a .so missing, or a .so loaded from a *different directory* or... a config file that is missing or... basically something right at the start.
by the time you get to "the fonts are not loaded"... well... you already know that, so you're clearly *not* looking for that...
... *UNLESS*....
the locations where each system is *looking* for the fonts is different. *that* is something you will want to know.
track and *interpret* the differences... don't just go "errr there are differences". work backwards for what the *important* differences are.
l.
) = 5 read(5, "<?xml version=\"1.0\"?>\n<!DOCTYPE "..., 8192) = 8192 read(5, "ol>true</bool></test>\n\t <family"..., 8192) = 8192 read(5, "rue</bool></test>\n\t <family>Ari"..., 8192) = 1737 read(5, "", 8192) = 0 close(5) = 0 -getrandom("\xb9\x62\x9e\xde\xf0\x67\x1e\xe4", 8, GRND_NONBLOCK) = 8 +getrandom("\xbd\xd6\x85\x56\x07\x99\x3b\x97", 8, GRND_NONBLOCK) = 8 +access("/etc/fonts/conf.d/31-cantarell.conf", R_OK) = 0 +stat("/etc/fonts/conf.d/31-cantarell.conf", {st_mode=S_IFREG|0644, st_size=440, ...}) = 0 +openat(AT_FDCWD, "/etc/fonts/conf.d/31-cantarell.conf", O_RDONLY|O_CLOEXEC) = 5 +read(5, "<?xml version="1.0" encoding="UT"..., 8192) = 440 +read(5, "", 8192) = 0 +close(5) = 0 +getrandom("\xbe\xfa\x92\x55\x6b\x7e\xf1\xef", 8, GRND_NONBLOCK) = 8 access("/etc/fonts/conf.d/31-linux-libertine.conf", R_OK) = 0 stat("/etc/fonts/conf.d/31-linux-libertine.conf", {st_mode=S_IFREG|0644, st_size=986, ...}) = 0 openat(AT_FDCWD, "/etc/fonts/conf.d/31-linux-libertine.conf", O_RDONLY|O_CLOEXEC) = 5 read(5, "<fontconfig>\n <alias>\n "..., 8192) = 986 read(5, "", 8192) = 0
third major difference: /etc/fonts/conf.d/31-cantarell.conf is loaded on the *EXACT* same system that reads /root/.config/fontconfig/rendering-options.conf (the exact same system that did NOT have a locale set
--- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 11:01 AM, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
read(5, "<?xml version='1.0'?>\n<!DOCTYPE "..., 8192) = 3364 read(5, "", 8192) = 0 close(5) = 0 -getrandom("\xd8\xfc\x79\x69\xc3\xed\x3f\x71", 8, GRND_NONBLOCK) = 8 -stat("/root/.config/fontconfig/rendering-options.conf", {st_mode=S_IFREG|0664, st_size=517, ...}) = 0 -stat("/root/.config/fontconfig/rendering-options.conf", {st_mode=S_IFREG|0664, st_size=517, ...}) = 0 -access("/root/.config/fontconfig/rendering-options.conf", R_OK) = 0 -stat("/root/.config/fontconfig/rendering-options.conf", {st_mode=S_IFREG|0664, st_size=517, ...}) = 0 -openat(AT_FDCWD, "/root/.config/fontconfig/rendering-options.conf", O_RDONLY|O_CLOEXEC) = 5 -read(5, "<?xml version=\"1.0\"?>\n<!DOCTYPE "..., 8192) = 517 -read(5, "", 8192) = 0 -close(5) = 0 -getrandom("\x20\xa3\xc2\x43\x63\x24\x77\xb2", 8, GRND_NONBLOCK) = 8 -access("/root/.config/fontconfig/rendering-options.conf", R_OK) = 0 +getrandom("\xb3\x97\xce\xa5\xef\x3f\xce\xd0", 8, GRND_NONBLOCK) = 8 +stat("/root/.config/fontconfig/rendering-options.conf", 0x7fff190f8e30) = -1 ENOENT (No such file or directory) +stat("/root/.config/fontconfig/rendering-options.conf", 0x7fff190f8e30) = -1 ENOENT (No such file or directory) +access("/root/.config/fontconfig/rendering-options.conf", R_OK) = -1 ENOENT (No such file or directory) +access("/root/.config/fontconfig/rendering-options.conf", R_OK) = -1 ENOENT (No such file or directory)
second major difference: /root/.config/fontconfig/rendering-options.conf is not empty on one of the systems.
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 11:00 AM, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
open("/usr/share/locale/locale.alias", O_RDONLY|O_CLOEXEC) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=2939, ...}) = 0 -read(3, "# Locale name alias data base.\n#"..., 4096) = 2939 -read(3, "", 4096) = 0 +read(3, "# Locale name alias data base.\n#"..., 1024) = 1024 +read(3, "nd for the time being for\n# back"..., 1024) = 1024 +read(3, "59-1\ngalego\t\tgl_ES.ISO-8859-1\nga"..., 1024) = 891 +read(3, "", 1024) = 0 close(3) = 0
first major difference: locale not set.
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 10:56 AM, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
On Sun, Jun 3, 2018 at 9:27 AM, Felix Miata mrmazda@earthlink.net wrote:
deloptes composed on 2018-06-03 09:26 (UTC+0200):
Luke Kenneth Casson Leighton wrote:
you could instead put the two files side-by-side in xterms of identical height and do page-down page-down page-down instead.
there is kdiff3 for that
It does a nice job of showing off differences. I see lots of differences related to which fonts are or are not installed, and the different hostnames and various addresses, but I fail to ID any missing bit to ID why 760's 'tdecmshell fonts' fails and 780's succeeds.
you'll be looking for the first "major" obvious difference. "the font is missing" is f****g useless: you already know that. it will be something like a directory name, or a .so missing, or a .so loaded from a *different directory* or... a config file that is missing or... basically something right at the start.
by the time you get to "the fonts are not loaded"... well... you already know that, so you're clearly *not* looking for that...
... *UNLESS*....
the locations where each system is *looking* for the fonts is different. *that* is something you will want to know.
track and *interpret* the differences... don't just go "errr there are differences". work backwards for what the *important* differences are.
l.
read(4, "\4\374\2\374\7\0\0\0\250\1\0\0\0\0\0\0@\0\0\0\0\0\0\0X\0\0\0\0\0\0\0"..., 424) = 424 close(4) = 0 -stat("/usr/X11R6/lib/X11/fonts", 0x7fffb06ec980) = -1 ENOENT (No such file or directory) -stat("/usr/X11R6/lib/X11/fonts", 0x7fffb06ec9f0) = -1 ENOENT (No such file or directory) -stat("/opt/kde3/share/fonts", 0x7fffb06ec980) = -1 ENOENT (No such file or directory) -stat("/opt/kde3/share/fonts", 0x7fffb06ec9f0) = -1 ENOENT (No such file or directory) -stat("/usr/local/share/fonts", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 +stat("/usr/X11R6/lib/X11/fonts", 0x7fff19101bb0) = -1 ENOENT (No such file or directory) +stat("/usr/X11R6/lib/X11/fonts", 0x7fff19101c20) = -1 ENOENT (No such file or directory) +stat("/opt/kde3/share/fonts", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +openat(AT_FDCWD, "/opt/kde3/share/fonts", O_RDONLY|O_CLOEXEC) = 4 +fstatfs(4, {f_type=EXT2_SUPER_MAGIC, f_bsize=1024, f_blocks=5655093, f_bfree=719784, f_bavail=428969, f_files=358400, f_ffree=251758, f_fsid={val=[1551756211, 225110432]}, f_namelen=255, f_frsize=1024, f_flags=ST_VALID|ST_NOATIME}) = 0 +close(4) = 0 +openat(AT_FDCWD, "/var/cache/fontconfig//cf6c88e680607f2ab796171745f068a4-x86_64.cache-7", O_RDONLY|O_CLOEXEC) = 4 +fstat(4, {st_mode=S_IFREG|0644, st_size=104, ...}) = 0 +fstatfs(4, {f_type=EXT2_SUPER_MAGIC, f_bsize=1024, f_blocks=5655093, f_bfree=719783, f_bavail=428968, f_files=358400, f_ffree=251758, f_fsid={val=[1551756211, 225110432]}, f_namelen=255, f_frsize=1024, f_flags=ST_VALID|ST_NOATIME}) = 0 +read(4, "\4\374\2\374\7\0\0\0h\0\0\0\0\0\0\0@\0\0\0\0\0\0\0X\0\0\0\0\0\0\0"..., 104) = 104 +close(4) = 0 +stat("/usr/local/share/fonts", {st_mode=S_IFDIR|S_ISGID|0775, st_size=4096, ...}) = 0 openat(AT_FDCWD, "/usr/local/share/fonts", O_RDONLY|O_CLOEXEC) = 4
fourth major difference: on the first system, /opt/kde3/share/fonts does not exist.
my point is: you can't just go "um the files look different, there appears to be some stuff about fonts", you have to *LOOK* at the file differences, see what the system calls are, and be discerning.
if you're not familiar with this concept see "Knowledge Type 3" here: https://en.wikipedia.org/wiki/Advaita_Vedanta#Epistemology_-_ways_of_knowing
l.
120) = 120 close(4) = 0 openat(AT_FDCWD, "/usr/share/X11/locale/locale.alias", O_RDONLY) = 4 fstat(4, {st_mode=S_IFREG|0644, st_size=81760, ...}) = 0 -read(4, "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n"..., 4096) = 4096 -read(4, "IN.UTF-8\nbo_IN.utf8\t\t\t\t\tbo_IN.UT"..., 4096) = 4096 -read(4, "o\t\t\t\t\tde_DE.ISO8859-15\nde_DE.885"..., 4096) = 4096
fifth major difference: /usr/share/X11/local/locale.alias appears to contain far more lines on the second system than on the first.
access("/usr/share/X11/locale/en_US.UTF-8/XLC_LOCALE", R_OK) = 0 openat(AT_FDCWD, "/usr/share/X11/locale/en_US.UTF-8/XLC_LOCALE", O_RDONLY) = 4 fstat(4, {st_mode=S_IFREG|0644, st_size=4254, ...}) = 0 -read(4, "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n"..., 4096) = 4096 -read(4, "R\n}\n \n#\tcs20 class\ncs15\t{\n\tside\t"..., 4096) = 158 -read(4, "", 4096) = 0 +read(4, "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n"..., 1024) = 1024 +read(4, "\tISO8859-7:GR\n\t}\n}\n#\tfs9 class ("..., 1024) = 1024 +read(4, "24 class\nfs16\t{\n\tcharset\t{\n\t\tnam"..., 1024) = 1024 +read(4, " GR\n length "..., 1024) = 1024 +read(4, "R\n}\n \n#\tcs20 class\ncs15\t{\n\tside\t"..., 1024) = 158 +read(4, "", 1024) = 0 close(4) = 0
the en_US.UTF-8 LC_LOCALE is also different.
-lstat("/root/.trinity/share/config", {st_mode=S_IFDIR|0700, st_size=4096, ...}) = 0 -lstat("/opt", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 -lstat("/opt/trinity", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 -lstat("/opt/trinity/share", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0 -lstat("/opt/trinity/share/config", 0x7fffb06eb230) = -1 ENOENT (No such file or directory) -access("/opt/trinity/share/config", F_OK) = -1 ENOENT (No such file or directory) -stat("/root/.trinity/share/config/kdeglobals", {st_mode=S_IFREG|0600, st_size=7834, ...}) = 0 -stat("/etc/trinity/kdeglobals", 0x7fffb06ec1f0) = -1 ENOENT (No such file or directory) +lstat("/root/.trinity", {st_mode=S_IFDIR|0700, st_size=1024, ...}) = 0 +lstat("/root/.trinity/share", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +lstat("/root/.trinity/share/config", {st_mode=S_IFDIR|0755, st_size=3072, ...}) = 0 +lstat("/opt", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +lstat("/opt/trinity", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +lstat("/opt/trinity/share", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +lstat("/opt/trinity/share/config", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +access("/opt/trinity/share/config", F_OK) = 0 +lstat("/opt/trinity/share/config", {st_mode=S_IFDIR|0755, st_size=1024, ...}) = 0 +stat("/root/.trinity/share/config/kdeglobals", {st_mode=S_IFREG|0600, st_size=8624, ...}) = 0 +stat("/etc/trinity/kdeglobals", 0x7fff19101420) = -1 ENOENT (No such file or directory) +stat("/opt/trinity/share/config/kdeglobals", 0x7fff19101420) = -1 ENOENT (No such file or directory) access("/etc/tderc", R_OK) = -1 ENOENT (No such file or directory) -stat("/root/.trinity/share/config/system.kdeglobals", 0x7fffb06ec1f0) = -1 ENOENT (No such file or directory) -stat("/etc/trinity/system.kdeglobals", 0x7fffb06ec1f0) = -1 ENOENT (No such file or directory) +stat("/root/.trinity/share/config/system.kdeglobals", 0x7fff19101420) = -1 ENOENT (No such file or directory) +stat("/etc/trinity/system.kdeglobals", 0x7fff19101420) = -1 ENOENT (No such file or directory) +stat("/opt/trinity/share/config/system.kdeglobals", 0x7fff19101420) = -1 ENOENT (No such file or directory) openat(AT_FDCWD, "/root/.trinity/share/config/kdeglobals", O_RDONLY) = 11
seventh major difference: the 2nd system is looking for and finds a /opt/trinity/share/config directory... the first does not have that directory (line 5 of the above):
-lstat("/opt/trinity/share/config", 0x7fffb06eb230) = -1 ENOENT (No such file or directory)
how were each of these systems created? they're clearly not the same. i'd strongly suggest doing a dpkg -L on each (or dpkg-get-selections), a find / (with excluding of /dev, /proc etc - see updatedb's cron job for an example) on each and then do a diff on those.
l.
access("/root/.trinity/share/locale/POSIX/LC_MESSAGES/tdecmshell.mo", R_OK) = -1 ENOENT (No such file or directory) access("/opt/trinity/share/locale/POSIX/LC_MESSAGES/tdecmshell.mo", R_OK) = -1 ENOENT (No such file or directory) -access("/usr/share/locale-langpack/POSIX/LC_MESSAGES/tdecmshell.mo", R_OK) = -1 ENOENT (No such file or directory) access("/root/.trinity/share/locale-bundle/POSIX/LC_MESSAGES/tdecmshell.mo", R_OK) = -1 ENOENT (No such file or directory)
9th major difference: there are signs that the directory /usr/share/locale-langpack does not exist:
+access("/usr/share/locale-langpack", F_OK) = -1 ENOENT (No such file or directory)
also, one of the systems appears to have been set to plastik style, the other to keramik
at that point the differences becomes significant (line 2603) because the styles do radically different things. one is looking for /etc/tqt3/qtrc and finds it, the other doesn't look for it.
l.
Luke Kenneth Casson Leighton composed on 2018-06-03 11:22 (UTC+0100):
9th major difference: there are signs that the directory /usr/share/locale-langpack does not exist:
+access("/usr/share/locale-langpack", F_OK) = -1 ENOENT (No such file or directory)
also, one of the systems appears to have been set to plastik style, the other to keramik
at that point the differences becomes significant (line 2603) because the styles do radically different things. one is looking for /etc/tqt3/qtrc and finds it, the other doesn't look for it.
I recognized all those fontconfig/installed fonts/not installed fonts to be irrelevant. The fonts unit worked find via kcontrol.
I've updated https://bugs.pearsoncomputing.net/show_bug.cgi?id=2821 with the cause: having yast2-fonts installed breaks the finding of module fonts.
--- crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
On Sun, Jun 3, 2018 at 1:31 PM, Felix Miata mrmazda@earthlink.net wrote:
Luke Kenneth Casson Leighton composed on 2018-06-03 11:22 (UTC+0100):
9th major difference: there are signs that the directory /usr/share/locale-langpack does not exist:
+access("/usr/share/locale-langpack", F_OK) = -1 ENOENT (No such file or directory)
also, one of the systems appears to have been set to plastik style, the other to keramik
at that point the differences becomes significant (line 2603) because the styles do radically different things. one is looking for /etc/tqt3/qtrc and finds it, the other doesn't look for it.
I recognized all those fontconfig/installed fonts/not installed fonts to be irrelevant. The fonts unit worked find via kcontrol.
I've updated https://bugs.pearsoncomputing.net/show_bug.cgi?id=2821 with the cause: having yast2-fonts installed breaks the finding of module fonts.
interesting. is that reproducible on both systems (add yast2-fonts on each, remove yast2-fonts on each). it would be worthwhile adding strace logs for both systems, before and after yast2-fonts installed.
l.
Luke Kenneth Casson Leighton composed on 2018-06-03 21:19 (UTC+0100):
Felix Miata wrote:
I've updated https://bugs.pearsoncomputing.net/show_bug.cgi?id=2821 with the cause: having yast2-fonts installed breaks the finding of module fonts.
interesting. is that reproducible on both systems (add yast2-fonts on each, remove yast2-fonts on each). it would be worthwhile adding strace logs for both systems, before and after yast2-fonts installed.
Bug updated again with easy third host reproduction. I brought this up on https://lists.opensuse.org/opensuse-kde3/2018-06/msg00000.html but more likely it might do any good on the yast2-devel list, which if it produces no response likely I'll do.
Felix Miata composed on 2018-06-03 17:57 (UTC-0400):
Luke Kenneth Casson Leighton composed on 2018-06-03 21:19 (UTC+0100):
Felix Miata wrote:
I've updated https://bugs.pearsoncomputing.net/show_bug.cgi?id=2821 with the cause: having yast2-fonts installed breaks the finding of module fonts.
interesting. is that reproducible on both systems (add yast2-fonts on each, remove yast2-fonts on each). it would be worthwhile adding strace logs for both systems, before and after yast2-fonts installed.
Bug updated again with easy third host reproduction. I brought this up on https://lists.opensuse.org/opensuse-kde3/2018-06/msg00000.html but more likely it might do any good on the yast2-devel list, which if it produces no response likely I'll do.
Since this also happens on Tumbleweed, I filed a downstream bug: https://bugzilla.opensuse.org/show_bug.cgi?id=1095704
-
deloptes -
Felix Miata -
Luke Kenneth Casson Leighton