Skip to content

Commit a76ccd0

Browse files
chore: Add debug log for the rapidocr resolution
Signed-off-by: Nikos Livathinos <nli@zurich.ibm.com>
1 parent 28f7ee9 commit a76ccd0

1 file changed

Lines changed: 20 additions & 12 deletions

File tree

docling/models/stages/ocr/rapid_ocr_model.py

Lines changed: 20 additions & 12 deletions
Original file line numberDiff line numberDiff line change
@@ -80,22 +80,30 @@ def _resolve_rapidocr(lang: list[str], backend: str) -> "tuple[OCRVersion, str]"
8080
aliased = COMMON_LANG_ALIASES.get(code, code)
8181

8282
if aliased in PP_OCRV6_LANGS:
83-
return OCRVersion.PPOCRV6, aliased
84-
85-
if backend == "torch":
86-
if aliased in _PPOCRV4_LANGS:
87-
return OCRVersion.PPOCRV4, aliased
83+
version = OCRVersion.PPOCRV6
84+
elif backend == "torch":
85+
if aliased not in _PPOCRV4_LANGS:
86+
raise ValueError(
87+
f"RapidOCR torch backend does not support language {langs[0]!r}. "
88+
f"Supported: {sorted(PP_OCRV6_LANGS | _PPOCRV4_LANGS)}."
89+
)
90+
version = OCRVersion.PPOCRV4
91+
elif aliased in _PPOCRV5_LANGS:
92+
version = OCRVersion.PPOCRV5
93+
else:
8894
raise ValueError(
89-
f"RapidOCR torch backend does not support language {langs[0]!r}. "
90-
f"Supported: {sorted(PP_OCRV6_LANGS | _PPOCRV4_LANGS)}."
95+
f"RapidOCR {backend} backend does not support language {langs[0]!r}. "
96+
f"Supported: {sorted(PP_OCRV6_LANGS | _PPOCRV5_LANGS)}."
9197
)
9298

93-
if aliased in _PPOCRV5_LANGS:
94-
return OCRVersion.PPOCRV5, aliased
95-
raise ValueError(
96-
f"RapidOCR {backend} backend does not support language {langs[0]!r}. "
97-
f"Supported: {sorted(PP_OCRV6_LANGS | _PPOCRV5_LANGS)}."
99+
_log.debug(
100+
"RapidOCR resolved lang=%r backend=%r -> version=%s rec_lang=%r",
101+
lang,
102+
backend,
103+
version.value,
104+
aliased,
98105
)
106+
return version, aliased
99107

100108

101109
def _download_if_missing(url: str, dest: Path, *, force: bool, progress: bool) -> Path:

0 commit comments

Comments
 (0)