Spaces:
Sleeping
Sleeping
minor fix on "Python"
Browse files
app.py
CHANGED
|
@@ -5,15 +5,29 @@ import urllib
|
|
| 5 |
import requests
|
| 6 |
|
| 7 |
# FEATURE: get wikipedia codes, e.g. from https://en.wikipedia.org/wiki/List_of_Wikipedias or https://meta.wikimedia.org/wiki/List_of_Wikipedias, some of which are nonstandard. Then output f"{code}.wikipedia.org"
|
|
|
|
| 8 |
# TODO: fix 'knh', it has an empty ISO section. Turns out some languages only have 639-3
|
| 9 |
# TODO: add in some nice things from https://docs.streamlit.io/library/cheatsheet like error codes and status messages.
|
| 10 |
# TODO: add in vachan search even if lang not found
|
| 11 |
# TODO: results from glottolog even if none from others
|
|
|
|
| 12 |
things_to_test = [
|
| 13 |
"knh", # deprecated code on ISO
|
| 14 |
"khn", # only has 639-3 on ISO
|
| 15 |
"xxx", # no such code on ISO or glottolog
|
| 16 |
"Chinese", # Vachan struggles.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
]
|
| 18 |
|
| 19 |
|
|
@@ -76,26 +90,26 @@ def main():
|
|
| 76 |
return
|
| 77 |
|
| 78 |
if langtext.lower() == "python":
|
| 79 |
-
st.success("[Python is the best language!(https://www.python.org/)")
|
| 80 |
return
|
| 81 |
|
| 82 |
# TODO: st.code() for these "lookup in progress" outputs.
|
| 83 |
-
st.
|
| 84 |
|
| 85 |
if langcodes.tag_is_valid(langtext):
|
| 86 |
-
st.
|
| 87 |
else:
|
| 88 |
-
st.
|
| 89 |
|
| 90 |
|
| 91 |
try:
|
| 92 |
lang = langcodes.Language.get(langtext)
|
| 93 |
# st.write(f"{lang} is the BCP-47 tag.")
|
| 94 |
if "unknown" in lang.display_name().lower():
|
| 95 |
-
st.
|
| 96 |
lang = None
|
| 97 |
except langcodes.LanguageTagError as e:
|
| 98 |
-
st.
|
| 99 |
lang = None
|
| 100 |
|
| 101 |
|
|
@@ -104,16 +118,19 @@ def main():
|
|
| 104 |
try:
|
| 105 |
found = langcodes.find(langtext)
|
| 106 |
lang = found
|
| 107 |
-
st.
|
| 108 |
except LookupError as e:
|
| 109 |
-
st.
|
| 110 |
-
st.
|
| 111 |
st.write(f"* You can also try https://r12a.github.io/app-subtags/")
|
| 112 |
st.write(f"* Or possibly https://glottolog.org/glottolog?search={urllib.parse.quote(langtext)}")
|
| 113 |
lang = None
|
| 114 |
|
| 115 |
|
|
|
|
| 116 |
|
|
|
|
|
|
|
| 117 |
|
| 118 |
#st.write(f"langcodes found the following tag: {type(found)}") # a Language object
|
| 119 |
if lang is not None:
|
|
@@ -121,12 +138,12 @@ def main():
|
|
| 121 |
b_variant = lang.to_alpha3(variant='B')
|
| 122 |
t_variant = lang.to_alpha3(variant='T')
|
| 123 |
broader_tags = lang.broader_tags()
|
| 124 |
-
|
| 125 |
standardized_tag = langcodes.standardize_tag(lang)
|
| 126 |
-
|
| 127 |
|
| 128 |
|
| 129 |
-
st.write(f"## Results: probably use '{standardized_tag}'")
|
| 130 |
# TODO: make a results dictionary so it's easy to copy-paste?
|
| 131 |
st.write(f"Best-match BCP-47 tag for '{langtext}', according to the langcodes library: {lang}")
|
| 132 |
st.write(f"Breakdown of tag components:")
|
|
@@ -143,15 +160,8 @@ def main():
|
|
| 143 |
|
| 144 |
st.write(f"### Language Subtag Search Tool")
|
| 145 |
st.write(f"Try also: https://r12a.github.io/app-subtags/?lookup={lang}, which will likely have links to Ethnologue, Wikipedia, and Character usage. You can also try searching for '{langtext}' there!")
|
| 146 |
-
|
| 147 |
-
|
| 148 |
-
if languoid_id:
|
| 149 |
-
st.write(f"**Glottolog Languoid ID:** Searching for '{langtext}' on Glottolog returns the following 'languoid ID': [{languoid_id}](https://glottolog.org/resource/languoid/id/{languoid_id})")
|
| 150 |
-
st.write(f"https://glottolog.org/glottolog?search={t_variant} may be also of interest, with links to various resources including WALS, Wikidata, Odin, and OLAC. ")
|
| 151 |
-
if t_variant != b_variant:
|
| 152 |
-
st.write(f"If that doesn't work, try https://glottolog.org/glottolog?search={b_variant}, or put in a [custom search query](https://glottolog.org/glottolog)")
|
| 153 |
-
st.write(f"https://glottolog.org/glottolog?search={urllib.parse.quote(langtext)} may pull up something as well.")
|
| 154 |
-
|
| 155 |
st.write("### Older / Related Codes")
|
| 156 |
|
| 157 |
st.write(f"ISO 639-3 'alpha3' code, 'terminology' or 'T' variant (deprecated): {t_variant}")
|
|
@@ -172,11 +182,22 @@ def main():
|
|
| 172 |
elif b_obsolete_codes:
|
| 173 |
st.write(f"Obsolete codes from previous ISO-639 iterations, pulled from https://iso639-3.sil.org/code/{b_variant}:")
|
| 174 |
st.write(b_obsolete_codes)
|
| 175 |
-
|
| 176 |
-
|
| 177 |
-
|
| 178 |
-
|
| 179 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
|
| 181 |
|
| 182 |
if __name__ == "__main__":
|
|
|
|
| 5 |
import requests
|
| 6 |
|
| 7 |
# FEATURE: get wikipedia codes, e.g. from https://en.wikipedia.org/wiki/List_of_Wikipedias or https://meta.wikimedia.org/wiki/List_of_Wikipedias, some of which are nonstandard. Then output f"{code}.wikipedia.org"
|
| 8 |
+
# Big TODO: collate all the results into a big dictionary? Then display that. Reduces if statements?
|
| 9 |
# TODO: fix 'knh', it has an empty ISO section. Turns out some languages only have 639-3
|
| 10 |
# TODO: add in some nice things from https://docs.streamlit.io/library/cheatsheet like error codes and status messages.
|
| 11 |
# TODO: add in vachan search even if lang not found
|
| 12 |
# TODO: results from glottolog even if none from others
|
| 13 |
+
# TODO: check glottolog results to see if they find anything!
|
| 14 |
things_to_test = [
|
| 15 |
"knh", # deprecated code on ISO
|
| 16 |
"khn", # only has 639-3 on ISO
|
| 17 |
"xxx", # no such code on ISO or glottolog
|
| 18 |
"Chinese", # Vachan struggles.
|
| 19 |
+
"Mandarin", # Vachan struggles.
|
| 20 |
+
"zh-CN",
|
| 21 |
+
"Chinese",
|
| 22 |
+
"zh-Latn-pinyin",
|
| 23 |
+
"en-Latn-US",
|
| 24 |
+
"en",
|
| 25 |
+
"English",
|
| 26 |
+
"fr-CA",
|
| 27 |
+
"French (Canada)",
|
| 28 |
+
"français",
|
| 29 |
+
"法语",
|
| 30 |
+
"", # empty string
|
| 31 |
]
|
| 32 |
|
| 33 |
|
|
|
|
| 90 |
return
|
| 91 |
|
| 92 |
if langtext.lower() == "python":
|
| 93 |
+
st.success("[Python is the best language!](https://www.python.org/)")
|
| 94 |
return
|
| 95 |
|
| 96 |
# TODO: st.code() for these "lookup in progress" outputs.
|
| 97 |
+
st.info("* Checking whether the tag is valid. That is, the language, script, territory, and variants (if present) are all tags that have meanings assigned by IANA.")
|
| 98 |
|
| 99 |
if langcodes.tag_is_valid(langtext):
|
| 100 |
+
st.info(f"* ...True! '{langtext}' parses meaningfully as a language tag according to IANA.")
|
| 101 |
else:
|
| 102 |
+
st.info(f"* ...False! '{langtext}' doesn't parse meaningfully as a language tag according to IANA, some of its subcomponents may be invalid or it might be a natural language description.")
|
| 103 |
|
| 104 |
|
| 105 |
try:
|
| 106 |
lang = langcodes.Language.get(langtext)
|
| 107 |
# st.write(f"{lang} is the BCP-47 tag.")
|
| 108 |
if "unknown" in lang.display_name().lower():
|
| 109 |
+
st.info(f"* Attempting to lookup the code directly gives us '{lang.display_name()}', attempting to search for it as a natural language string.")
|
| 110 |
lang = None
|
| 111 |
except langcodes.LanguageTagError as e:
|
| 112 |
+
st.info(f"* Could not lookup code directly, attempting to search for it as a natural language string.")
|
| 113 |
lang = None
|
| 114 |
|
| 115 |
|
|
|
|
| 118 |
try:
|
| 119 |
found = langcodes.find(langtext)
|
| 120 |
lang = found
|
| 121 |
+
st.success(f"* Natural language search found the following BCP-47 tag: {lang}")
|
| 122 |
except LookupError as e:
|
| 123 |
+
st.error("## Result: failure!")
|
| 124 |
+
st.error(f"Unable to look up language code. But all hope is not lost...")
|
| 125 |
st.write(f"* You can also try https://r12a.github.io/app-subtags/")
|
| 126 |
st.write(f"* Or possibly https://glottolog.org/glottolog?search={urllib.parse.quote(langtext)}")
|
| 127 |
lang = None
|
| 128 |
|
| 129 |
|
| 130 |
+
|
| 131 |
|
| 132 |
+
t_variant = None
|
| 133 |
+
|
| 134 |
|
| 135 |
#st.write(f"langcodes found the following tag: {type(found)}") # a Language object
|
| 136 |
if lang is not None:
|
|
|
|
| 138 |
b_variant = lang.to_alpha3(variant='B')
|
| 139 |
t_variant = lang.to_alpha3(variant='T')
|
| 140 |
broader_tags = lang.broader_tags()
|
| 141 |
+
|
| 142 |
standardized_tag = langcodes.standardize_tag(lang)
|
| 143 |
+
|
| 144 |
|
| 145 |
|
| 146 |
+
st.write(f"## BCP-47 Results: probably use '{standardized_tag}'")
|
| 147 |
# TODO: make a results dictionary so it's easy to copy-paste?
|
| 148 |
st.write(f"Best-match BCP-47 tag for '{langtext}', according to the langcodes library: {lang}")
|
| 149 |
st.write(f"Breakdown of tag components:")
|
|
|
|
| 160 |
|
| 161 |
st.write(f"### Language Subtag Search Tool")
|
| 162 |
st.write(f"Try also: https://r12a.github.io/app-subtags/?lookup={lang}, which will likely have links to Ethnologue, Wikipedia, and Character usage. You can also try searching for '{langtext}' there!")
|
| 163 |
+
|
| 164 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
st.write("### Older / Related Codes")
|
| 166 |
|
| 167 |
st.write(f"ISO 639-3 'alpha3' code, 'terminology' or 'T' variant (deprecated): {t_variant}")
|
|
|
|
| 182 |
elif b_obsolete_codes:
|
| 183 |
st.write(f"Obsolete codes from previous ISO-639 iterations, pulled from https://iso639-3.sil.org/code/{b_variant}:")
|
| 184 |
st.write(b_obsolete_codes)
|
| 185 |
+
|
| 186 |
+
st.write(f"### Glottolog")
|
| 187 |
+
languoid_id = try_retrieving_glottolog_id(langtext)
|
| 188 |
+
if languoid_id:
|
| 189 |
+
st.write(f"**Glottolog Languoid ID:** Searching for '{langtext}' on Glottolog returns the following 'languoid ID': [{languoid_id}](https://glottolog.org/resource/languoid/id/{languoid_id})")
|
| 190 |
+
# FIXME: fix this to display something if there's an ISO code to try
|
| 191 |
+
if t_variant:
|
| 192 |
+
st.write(f"https://glottolog.org/glottolog?search={t_variant} may be also of interest, with links to various resources including WALS, Wikidata, Odin, and OLAC. ")
|
| 193 |
+
if t_variant != b_variant:
|
| 194 |
+
st.write(f"If that doesn't work, try https://glottolog.org/glottolog?search={b_variant}, or put in a [custom search query](https://glottolog.org/glottolog)")
|
| 195 |
+
st.write(f"https://glottolog.org/glottolog?search={urllib.parse.quote(langtext)} may pull up something as well.")
|
| 196 |
+
|
| 197 |
+
results_from_vachan = try_searching_vachan_engine(langtext)
|
| 198 |
+
if results_from_vachan:
|
| 199 |
+
st.write("### Other potential matches, from [Vachan Engine](https://github.com/Bridgeconn/vachan-api/tree/version-2) (experimental)")
|
| 200 |
+
st.write(results_from_vachan)
|
| 201 |
|
| 202 |
|
| 203 |
if __name__ == "__main__":
|