Age | Commit message (Collapse) | Author |
|
|
|
Alternative to
https://github.com/lpereira/hardinfo/pull/418
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
x86 CPU:
* Remove (R), (TM), etc
* AMD: remove redundant "X2 Dual Core", "X4 Quad Core", etc
* Move vendor to front if not already
* Remove speed, as the actual speed is usually right next to it
* Remove "CPU", "APU", "Processor" to shorten
* Compress space
Intel GPU:
* Remove (R)
* Abbreviate "Generation": "Gen"
* Remove "Processor", "Controller", and "Device" to shorten
* Compress space
Some examples:
x86 CPU:
'AMD Turion(tm) 64 Mobile Technology ML-32' ---> 'AMD Turion 64 Mobile ML-32'
'Cyrix MediaGXtm MMXtm Enhanced' ---> 'Cyrix MediaGX MMX Enhanced'
'Transmeta(tm) Crusoe(tm) Processor TM5800' ---> 'Transmeta Crusoe TM5800'
'VIA Nano X2 L4350 @ 1.6+ GHz' ---> 'VIA Nano X2 L4350'
'AMD Athlon(tm) 64 X2 Dual-Core Processor TK-53' ---> 'AMD Athlon 64 X2 TK-53'
'Embedded AMD Opteron(tm) Processor 23KS EE' ---> 'AMD Embedded Opteron 23KS EE'
'Intel(R) Atom(TM) x5-Z8300 CPU @ 1.44GHz' ---> 'Intel Atom x5-Z8300'
'Intel(R) Pentium(R) III CPU - S 1400MHz' ---> 'Intel Pentium III - S'
'Dual Core AMD Opteron(tm) Processor 165' ---> 'AMD Dual Core Opteron 165'
'Genuine Intel(R) CPU T1350 @ 1.86GHz' ---> 'Intel T1350'
'AMD Phenom(tm) 9350e Quad-Core Processor' ---> 'AMD Phenom 9350e Quad-Core'
Intel GPU:
'Atom/Celeron/Pentium Processor x5-E8000/J3xxx/N3xxx Integrated Graphics Controller' ---> 'Atom/Celeron/Pentium x5-E8000/J3xxx/N3xxx Integrated Graphics'
'4th Generation Core Processor Family Integrated Graphics Controller' ---> '4th Gen Core Family Integrated Graphics'
'Mobile GM965/GL960 Integrated Graphics Controller (secondary)' ---> 'Mobile GM965/GL960 Integrated Graphics (secondary)'
'Mobile 915GM/GMS/910GML Express Graphics Controller' ---> 'Mobile 915GM/GMS/910GML Express Graphics'
Also:
Fix/replace the appf() and SEQ() that were peppered about.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
* more fields in struct Vendor.
* match_case -> match_rule and new match rule "exact". match_rule
remains compatible with match_case for old conf format.
* matches must be "whole word" matches so "Harmony" no longer hits
for "ARM".
* Parts outside of () are checked first, so "Foo (formerly Barly)"
matches Foo before Barly, even though Barly is longer and would
otherwise match first.
* vendor_list type (a blessed GSList*) and helper functions to manage
it.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
|
|
If lengths are equal then prefer
name_short > name > id_str.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
consistent
Following the old behavior of returning the passed-in string if not found.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
* Initialize strings to empty in `read_from_vendor_ids()`
* vendor.ids format reader clears all fields at `name`
* Added `name_short`, and `url_support` fields
* `vendor_cleanup()` function that frees `vendor_list`
* `vendor_match()` returns the whole `Vendor` data structure
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
Create a new, more easily maintained, file for vendor information.
Installs with hardinfo, but can be sync-ed if that function ever
gets restored.
Reading the old vendor.conf format is still fully supported, AFAIK.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
Fix a really stupid sort function I wrote for 60b9f3360930296c0f3e8b04672b8bf7468bedcb.
Now actually does what that commit claims to do.
I think I must have just sketched it out and then forgot to go back and correct it. It's
weird that it happened to fix the one result I was testing for.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
Build the configuration file path using g_get_user_config_dir() instead of g_get_home_dir().
|
|
* Rename struct members to be less ambiguous
* Sort the list by length of match_string, longest first,
so better matches match first
* Add flag to force case-sensitive matching so that two or
three-letter names don't match names that happen to have
those letters in them
Also, added ASUS and fixes #241.
(https://github.com/lpereira/hardinfo/issues/241)
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
x86/processor.c uses vendor_get_name() with the CPUID vendor
string as argument.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
* nVidia tweak
* Don't strdup() vendor_get_*() results, they return const and nobody's
freeing them anyway.
Signed-off-by: Burt P <pburt0@gmail.com>
|
|
|