Issue: Invalid literal for some Locales

[gonsolo]

I started two vlc instances, then try to sync with vlcsync, but the following error occurs:

Cannot connect to 127.0.0.42:40101 (pid=12437) socket, cause: invalid literal for int() with base 10: '256,0'. Skipping. Enable debug for more info. See --help.

[gonsolo]

The following hack works:

diff --git a/vlcsync/vlc.py b/vlcsync/vlc.py
index 552f29a..ad95d43 100644
--- a/vlcsync/vlc.py
+++ b/vlcsync/vlc.py
@@ -46,6 +46,8 @@ class Vlc:
     def volume(self) -> Optional[int]:
         vol = self.vlc_conn.cmd("volume")
         if vol.strip() != '':
+            locale.setlocale(locale.LC_ALL, locale.getlocale())
+            vol = locale.atof(vol)
             return int(vol)
         else:
             return None

[mrkeuz]

Thanks for contribution!

ok. Your first assumption was split(','), now locale. Why locale? We read output from another app (vlc) with another Locale AFIAK? How locale affected output from tcp socket?

IDK what mean each number here '256,0'. It just fraction of volume? Just on my machines it always integer.

Can you test? Just connect to vlc, change volume and see what volume returns.

➜ telnet 127.0.0.42 46489 #You port here from vlcsyc
Trying 127.0.0.42...
Connected to 127.0.0.42.
Escape character is '^]'.
VLC media player 3.0.9.2 Vetinari
Command Line Interface initialized. Type `help' for help.
> volume
0
> volume
120
>  

My always int as you see.

I assume put just:

>>> int(float(vol.replace(",",'.')))

Dirty but we cover all cases.

What OS and vlc version did you use?

[gonsolo]

ok. Your first assumption was split(','), now locale. Why locale? We read output from another app (vlc) with another Locale AFIAK? How locale affected output from tcp socket?

My LANG is de_DE.UTF-8, your is probably C.

IDK what mean each number here '256,0'. It just fraction of volume? Just on my machines it always integer.

It's probably a float.

Can you test? Just connect to vlc, change volume and see what volume returns.

Trying ::1...
Connected to localhost.
Escape character is '^]'.
VLC media player 3.0.21 Vetinari
Password: 
Welcome, Master
> volume
0,0

Dirty but we cover all cases.

I'd prefer a proper solution.

What OS and vlc version did you use?

Latest Arch, vlc 3.0.21.

[mrkeuz]

I'd prefer a proper solution.

ok. Agree. Will try to test with different locales.

Thanks for testing! 🚀

[mrkeuz]

In the end reproduced issue locally. Yes it float.
Decide just support both formats. I don't want to rely on to locale.

Motivation:
Just imagine if vlc and vlcsync run with different locales. By some reason.
I cannot get from vlc locale. At least it technically hard in cross-platform.
So simple - allow multi-format. Will work in all imagined cases.

Cheers!

[Et0h]

https://en.wikipedia.org/wiki/Decimal_separator does list a small number of other decimal separators such as · and '.

I don't know if any of these alternative decimal separators are ever used by VLC, but to accommodate all possible uses cases you could create a function which identifies any non-numeric character and treats it as a decimal point. This would be fine so long as there is no thousand separator.

VLC's common.lua just uses tostring(n):gsub(",", ".", 1) as part of tostring() function for decimals number using a dot as decimal separator "regardless of the system locale". That implies that the VLC developers worked on the basis that VLC will not be producing any decimal separator other than , or ..

[mrkeuz]

@Et0h

other decimal separators such as · and '.

Good point.

but to accommodate all possible uses cases

OK. I will think about how it can be implemented more carefully. And what we can do with this. As you say we can get only thousand delimiters for Int and it can be only one. In this case non-digit cannot be considered as decimal delimiter.

[mrkeuz]

WIP: Must support all locales, NOT only '.',','

[Ultraporing]

Hmh.. the vlc player always encodes all strings to POSIX/C/American decimal format. They got methods to convert always to US formatting.

We could just do the same. Here is the lua implementations from VLC for reference:

• https://github.com/videolan/vlc/blob/master/share/lua/modules/common.lua#L3

Their C implementation overwrites the strtof / strtod ect. methods reference:

• header: https://github.com/videolan/vlc/blob/master/include/vlc_charset.h
• implementation: https://github.com/videolan/vlc/blob/master/src/text/charset.c

Article with decimal separator examples and the codes: https://randombits.dev/articles/number-localization/intro

Either we use a library for the localization or need to hand roll the replacement. As far as I could find, VLC only replaces , with . in lua. And uses the locale methods for C.

My attempt at implementing in the screenshots.

import locale

class CNum(float):
    def __new__(cls: CNum, value: CNum | int | float | str) -> CNum:
        oInt: float = CNum.prepareInputArguments(value)
        instance: float = super().__new__(cls, oInt)
        return instance
    
    def to_vnc_int_str(self) -> str:
        locale.setlocale(locale.LC_NUMERIC, "C")
        return locale.format_string("%.0f", self)
    
    def to_vnc_float_str(self) -> str:
        locale.setlocale(locale.LC_NUMERIC, "C")
        return locale.format_string("%.1f", self)

    @staticmethod
    def prepareInputArguments(value: CNum | int | float | str) -> float:
        locale.setlocale(locale.LC_NUMERIC, "C")
        oInt: float = None
        
        if isinstance(value, int | float):
            oInt = value
        elif isinstance(value, str):
            cl = value.strip().replace(",", ".", 1).replace("\u066b", ".", 1)    
            oInt = locale.atof(cl)
        else:
            assert oInt != None, "Failed to localize value"
        
        return oInt

Screenshots

Grok's Wall of Text Table of locale delimiters ect

Shared Locales	Short Date Format	Short Time Format	Decimal Separator	Thousand Separator
en_US, en_CA, fil_PH, ceb_PH, chr_US, en_PR, en_AS, en_GU, en_MP, en_VI, en_UM	M/d/yy	h:mm a	.	,
en_GB, en_AU, en_NZ, en_IE, en_ZA, en_IN, en_PH, en_PK, en_MY, en_SG, en_HK, en_MO, en_IL, en_KE, en_TZ, en_UG, en_ZM, en_ZW, en_BW, en_LS, en_MW, en_NA, en_RW, en_SS, en_SZ, en_BB, en_BM, en_BS, en_BZ, en_DM, en_GD, en_GM, en_GY, en_JM, en_KN, en_LC, en_LR, en_MS, en_PG, en_SB, en_SC, en_SL, en_TC, en_TT, en_TV, en_VC, en_VG, en_VU, en_WS, en_FJ, en_FK, en_FM, en_GI, en_KI, en_KY, en_MT, en_MU, en_NF, en_NR, en_NU, en_PN, en_SH, en_TK, en_TO	d/M/yyyy	HH:mm	.	,
fr_FR, fr_BE, fr_CA, fr_CH, fr_LU, fr_MC, fr_BF, fr_BI, fr_BJ, fr_CD, fr_CF, fr_CG, fr_CI, fr_CM, fr_DJ, fr_DZ, fr_GA, fr_GF, fr_GN, fr_GP, fr_GQ, fr_HT, fr_KM, fr_MA, fr_MG, fr_ML, fr_MQ, fr_MR, fr_MU, fr_NC, fr_NE, fr_PF, fr_PM, fr_RE, fr_RW, fr_SC, fr_SN, fr_SY, fr_TD, fr_TG, fr_TN, fr_VU, fr_WF, fr_YT	d/M/y	HH:mm	,	space
de_DE, de_AT, de_CH, de_LI, de_LU, de_BE, da_DK, da_GL, sv_SE, sv_AX, sv_FI, no_NO, no_SJ, fi_FI, is_IS, fo_FO, kl_GL, ksh_DE, nds_DE, nds_NL, dsb_DE, hsb_DE, smn_FI, se_FI, se_NO, se_SE, nn_NO, nb_NO, nb_SJ	d.M.y	HH:mm	,	space
es_ES, es_AR, es_BO, es_CL, es_CO, es_CR, es_CU, es_DO, es_EC, es_GT, es_HN, es_MX, es_NI, es_PA, es_PE, es_PR, es_PY, es_SV, es_UY, es_VE, pt_PT, pt_AO, pt_CV, pt_GQ, pt_GW, pt_LU, pt_MO, pt_MZ, pt_ST, pt_TL, nl_NL, nl_AR, nl_AW, nl_BQ, nl_CW, nl_SR, nl_SX, fy_NL, ast_ES, gl_ES, ca_ES, eu_ES, oc_ES, oc_FR	d/M/y	HH:mm	,	space
it_IT, it_CH, it_SM, it_VA, ro_RO, ro_MD, hr_HR, hr_BA, sl_SI, bs_BA, bs_Latn_BA, bs_Cyrl_BA, sr_RS, sr_Latn_RS, sr_Cyrl_RS, sr_BA, sr_Latn_BA, sr_Cyrl_BA, sr_ME, sr_Latn_ME, sr_Cyrl_ME, sr_XK, sr_Latn_XK, sr_Cyrl_XK	d.M.y	HH:mm	,	.
pt_BR, es_BR, es_CL, es_CO, es_PE	d/M/y	HH:mm	,	.
ja_JP, ko_KR, ko_KP, zh_CN, zh_SG, zh_Hans_CN, zh_Hans_SG, zh_Hant_TW, zh_Hant_HK, zh_Hant_MO, yue_HK, yue_Hans_CN, yue_Hant_HK	y/M/d	HH:mm	.	,
hu_HU, lt_LT, lv_LV, et_EE, pl_PL, cs_CZ, sk_SK, bg_BG, uk_UA, ru_RU, ru_BY, ru_KG, ru_KZ, ru_MD, ru_UA, be_BY, mk_MK, sq_AL, sq_MK, sq_XK, ky_KG, tg_TJ, uz_Cyrl_UZ, tt_RU, ba_RU, cv_RU, os_GE, os_RU, kk_KZ, az_Cyrl_AZ	d.M.y	HH:mm	,	space
th_TH, id_ID, vi_VN, ms_MY, ms_BN, ms_ID, ms_SG, km_KH, lo_LA	d/M/y	HH:mm	.	,
hi_IN, bn_BD, bn_IN, ne_NP, ne_IN, gu_IN, pa_IN, pa_Guru_IN, pa_Arab_PK, mr_IN, as_IN, or_IN, kn_IN, ml_IN, te_IN, ta_IN, ta_LK, ta_MY, ta_SG, si_LK, my_MM, bo_CN, bo_IN, dz_BT, brx_IN, kok_IN, doi_IN, mai_IN, mni_IN, sat_IN, mni_Beng_IN, bpy_IN, xnr_IN, kxv_Latn_IN, kxv_Deva_IN, kxv_Orya_IN, kxv_Telu_IN	d-M-y	HH:mm	.	, (3-2 grouping)
tr_TR, el_GR, el_CY, ku_TR, az_Latn_AZ, uz_Latn_UZ, tk_TM	d.M.y	HH:mm	,	.
ar_SA, ar_AE, ar_BH, ar_DJ, ar_DZ, ar_EG, ar_EH, ar_ER, ar_IL, ar_IQ, ar_JO, ar_KM, ar_KW, ar_LB, ar_LY, ar_MA, ar_MR, ar_OM, ar_PS, ar_QA, ar_SD, ar_SO, ar_SS, ar_SY, ar_TD, ar_TN, ar_YE, fa_IR, fa_AF, lrc_IR, mzn_IR, bqi_IR, ps_AF, ps_PK, ckb_IR, ckb_IQ	d/M/y	h:mm a	.	,
he_IL	d.M.y	HH:mm	.	,
am_ET, wal_ET	d/M/y	h:mm a	.	,
cgg_UG, dav_KE, ebu_KE, kam_KE, ki_KE, luo_KE, mas_KE, mas_TZ, mer_KE, rof_TZ, rw_RW, saq_KE, teo_KE, teo_UG, vun_TZ, asa_TZ, bez_TZ, lag_TZ, sbp_TZ	d/M/y	HH:mm	.	,
jgo_CM, dua_CM, ewo_CM, bas_CM, mgo_CM, mua_CM, nmg_CM, ksf_CM, agq_CM, yav_CM	d/M/y	HH:mm	,	space
kab_DZ, shi_Tfng_MA, zgh_MA	d/M/y	HH:mm	,	space
luy_KE, nyo_UG, lu_CD, ln_CD, ln_CF, ln_CG, ln_AO	d/M/y	HH:mm	.	,
ff_GN, ff_CM, ff_MR, ff_SN, ff_Adlm_GN, ff_Latn_GN	d/M/y	HH:mm	.	,
ha_GH, ha_NE	d/M/y	HH:mm	.	,
ig_NG, yo_NG, yo_BJ	d/M/y	h:mm a	.	,
ak_GH, ee_GH, ee_TG	d/M/y	HH:mm	.	,
nyn_UG, lg_UG, xog_UG	d/M/y	HH:mm	.	,
nd_ZW, sn_ZW	d/M/y	HH:mm	.	,
rn_BI, sg_CF	d/M/y	HH:mm	.	,
to_TO	d/M/y	h:mm a	.	,
tn_BW, tn_ZA	d/M/y	HH:mm	.	,
ts_ZA	d/M/y	HH:mm	.	,
ve_ZA	d/M/y	HH:mm	.	,
vo_001	d.M.y	HH:mm	.	,
yrl_BR, yrl_CO, yrl_VE	d/M/y	HH:mm	.	,
za_CN	d/M/y	HH:mm	.	,
zu_ZA	d/M/y	h:mm a	.	,