Skip to content

Commit 1aa99f7

Browse files
stainless-app[bot]stainless-app[bot]
committed
feat(api): api update
1 parent 5787a13 commit 1aa99f7

5 files changed

Lines changed: 300 additions & 98 deletions

File tree

‎.stats.yml‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
11
configured_endpoints: 30
2-
openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/context-dev/context.dev-c90f13e3cf4b9ae6a1d8f22c0e7285ab12d412d3a659a51a247a97adc7f08121.yml
3-
openapi_spec_hash: bf7879be38ebf3397a939c80b261a47a
2+
openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/context-dev/context.dev-fb56935a194e69348fecd985f7cf8b249795b46af2fb32f6c5c8ef648cf10c15.yml
3+
openapi_spec_hash: 7260a560474283b7ad6ac5d426058ac9
44
config_hash: daabb160675d86b354711da1e77e5129

‎README.md‎

Lines changed: 2 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -133,14 +133,8 @@ from context.dev import ContextDev
133133

134134
client = ContextDev()
135135

136-
response = client.web.extract(
137-
schema={
138-
"type": "bar",
139-
"properties": "bar",
140-
"required": "bar",
141-
"additionalProperties": "bar",
142-
},
143-
url="https://example.com",
136+
response = client.parse.handle(
137+
body=b"Example data",
144138
pdf={},
145139
)
146140
print(response.pdf)

‎src/context/dev/resources/parse.py‎

Lines changed: 179 additions & 50 deletions
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@
33
from __future__ import annotations
44

55
import os
6+
from typing_extensions import Literal
67

78
import httpx
89

@@ -59,14 +60,83 @@ def handle(
5960
self,
6061
body: FileContent | BinaryTypes,
6162
*,
62-
base_url: str | Omit = omit,
63-
extension: str | Omit = omit,
64-
filename: str | Omit = omit,
63+
extension: Literal[
64+
"txt",
65+
"text",
66+
"md",
67+
"markdown",
68+
"html",
69+
"htm",
70+
"xhtml",
71+
"xml",
72+
"rss",
73+
"atom",
74+
"csv",
75+
"tsv",
76+
"yaml",
77+
"yml",
78+
"py",
79+
"java",
80+
"js",
81+
"jsx",
82+
"mjs",
83+
"cjs",
84+
"json",
85+
"jsonl",
86+
"ndjson",
87+
"php",
88+
"sh",
89+
"bash",
90+
"zsh",
91+
"fish",
92+
"rb",
93+
"ts",
94+
"tsx",
95+
"rtf",
96+
"srt",
97+
"css",
98+
"scss",
99+
"less",
100+
"styl",
101+
"sass",
102+
"svg",
103+
"pdf",
104+
"docx",
105+
"doc",
106+
"xlsx",
107+
"xlsm",
108+
"xlsb",
109+
"xltx",
110+
"xltm",
111+
"xls",
112+
"pptx",
113+
"pptm",
114+
"ppsx",
115+
"ppsm",
116+
"potx",
117+
"potm",
118+
"ppt",
119+
"pps",
120+
"pot",
121+
"jpg",
122+
"jpeg",
123+
"jpe",
124+
"png",
125+
"gif",
126+
"bmp",
127+
"tiff",
128+
"tif",
129+
"webp",
130+
"ppm",
131+
"pbm",
132+
"pgm",
133+
"pnm",
134+
]
135+
| Omit = omit,
65136
include_images: bool | Omit = omit,
66137
include_links: bool | Omit = omit,
67138
ocr: bool | Omit = omit,
68-
pdf_end: int | Omit = omit,
69-
pdf_start: int | Omit = omit,
139+
pdf: parse_handle_params.Pdf | Omit = omit,
70140
shorten_base64_images: bool | Omit = omit,
71141
use_main_content_only: bool | Omit = omit,
72142
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -78,30 +148,28 @@ def handle(
78148
) -> ParseHandleResponse:
79149
"""
80150
Converts raw text, source code, web/data, PDF, Microsoft Office, and image bytes
81-
into LLM-usable Markdown.
151+
into LLM-usable Markdown. The base request costs 1 credit. When OCR runs
152+
(requires ocr=true), the entire call costs 5 credits; ocr=true requests where no
153+
OCR ends up running still cost 1 credit.
82154
83155
Args:
84-
base_url: Optional HTTP(S) source document URL used to resolve relative links and image
85-
references. Relative references remain relative when omitted.
86-
87-
extension: Optional file extension hint, such as pdf, docx, xlsx, pptx, html, json, csv,
88-
md, py, rtf, jpg, png, or txt.
89-
90-
filename: Optional filename hint used to infer the extension when extension is omitted.
156+
extension: Optional file extension hint. Case-insensitive; a leading dot is accepted (e.g.
157+
".pdf").
91158
92159
include_images: Include image references in Markdown output
93160
94161
include_links: Preserve hyperlinks in Markdown output
95162
96-
ocr: When true for PDF inputs, detect and OCR images embedded in the selected pages,
97-
inserting recognized text at each image's position in page reading order while
98-
preserving the PDF text layer. pdfStart/pdfEnd limit the inclusive page range.
99-
This is separate from automatic scanned-PDF OCR fallback.
100-
101-
pdf_end: Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
102-
Must be greater than or equal to pdfStart when both are provided.
163+
ocr: Gates all OCR. When true, PDFs get embedded-image OCR (recognized text inserted
164+
at each image's position in page reading order, preserving the text layer;
165+
pdf.start/pdf.end limit the page range), scanned PDFs with no text layer get
166+
full-document OCR, and raster images get their visible text transcribed. When
167+
false, no OCR runs: scanned PDFs may yield no content and images return only
168+
format/dimension metadata. Calls where OCR actually runs cost 5 credits instead
169+
of 1.
103170
104-
pdf_start: First 1-based PDF page to parse. When omitted, parsing starts at the first page.
171+
pdf: PDF page-range controls. Use start/end to limit parsing (and OCR when ocr=true)
172+
to an inclusive 1-based page range.
105173
106174
shorten_base64_images: Shorten base64-encoded image data in the Markdown output
107175
@@ -126,14 +194,11 @@ def handle(
126194
timeout=timeout,
127195
query=maybe_transform(
128196
{
129-
"base_url": base_url,
130197
"extension": extension,
131-
"filename": filename,
132198
"include_images": include_images,
133199
"include_links": include_links,
134200
"ocr": ocr,
135-
"pdf_end": pdf_end,
136-
"pdf_start": pdf_start,
201+
"pdf": pdf,
137202
"shorten_base64_images": shorten_base64_images,
138203
"use_main_content_only": use_main_content_only,
139204
},
@@ -168,14 +233,83 @@ async def handle(
168233
self,
169234
body: FileContent | AsyncBinaryTypes,
170235
*,
171-
base_url: str | Omit = omit,
172-
extension: str | Omit = omit,
173-
filename: str | Omit = omit,
236+
extension: Literal[
237+
"txt",
238+
"text",
239+
"md",
240+
"markdown",
241+
"html",
242+
"htm",
243+
"xhtml",
244+
"xml",
245+
"rss",
246+
"atom",
247+
"csv",
248+
"tsv",
249+
"yaml",
250+
"yml",
251+
"py",
252+
"java",
253+
"js",
254+
"jsx",
255+
"mjs",
256+
"cjs",
257+
"json",
258+
"jsonl",
259+
"ndjson",
260+
"php",
261+
"sh",
262+
"bash",
263+
"zsh",
264+
"fish",
265+
"rb",
266+
"ts",
267+
"tsx",
268+
"rtf",
269+
"srt",
270+
"css",
271+
"scss",
272+
"less",
273+
"styl",
274+
"sass",
275+
"svg",
276+
"pdf",
277+
"docx",
278+
"doc",
279+
"xlsx",
280+
"xlsm",
281+
"xlsb",
282+
"xltx",
283+
"xltm",
284+
"xls",
285+
"pptx",
286+
"pptm",
287+
"ppsx",
288+
"ppsm",
289+
"potx",
290+
"potm",
291+
"ppt",
292+
"pps",
293+
"pot",
294+
"jpg",
295+
"jpeg",
296+
"jpe",
297+
"png",
298+
"gif",
299+
"bmp",
300+
"tiff",
301+
"tif",
302+
"webp",
303+
"ppm",
304+
"pbm",
305+
"pgm",
306+
"pnm",
307+
]
308+
| Omit = omit,
174309
include_images: bool | Omit = omit,
175310
include_links: bool | Omit = omit,
176311
ocr: bool | Omit = omit,
177-
pdf_end: int | Omit = omit,
178-
pdf_start: int | Omit = omit,
312+
pdf: parse_handle_params.Pdf | Omit = omit,
179313
shorten_base64_images: bool | Omit = omit,
180314
use_main_content_only: bool | Omit = omit,
181315
# Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -187,30 +321,28 @@ async def handle(
187321
) -> ParseHandleResponse:
188322
"""
189323
Converts raw text, source code, web/data, PDF, Microsoft Office, and image bytes
190-
into LLM-usable Markdown.
324+
into LLM-usable Markdown. The base request costs 1 credit. When OCR runs
325+
(requires ocr=true), the entire call costs 5 credits; ocr=true requests where no
326+
OCR ends up running still cost 1 credit.
191327
192328
Args:
193-
base_url: Optional HTTP(S) source document URL used to resolve relative links and image
194-
references. Relative references remain relative when omitted.
195-
196-
extension: Optional file extension hint, such as pdf, docx, xlsx, pptx, html, json, csv,
197-
md, py, rtf, jpg, png, or txt.
198-
199-
filename: Optional filename hint used to infer the extension when extension is omitted.
329+
extension: Optional file extension hint. Case-insensitive; a leading dot is accepted (e.g.
330+
".pdf").
200331
201332
include_images: Include image references in Markdown output
202333
203334
include_links: Preserve hyperlinks in Markdown output
204335
205-
ocr: When true for PDF inputs, detect and OCR images embedded in the selected pages,
206-
inserting recognized text at each image's position in page reading order while
207-
preserving the PDF text layer. pdfStart/pdfEnd limit the inclusive page range.
208-
This is separate from automatic scanned-PDF OCR fallback.
209-
210-
pdf_end: Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
211-
Must be greater than or equal to pdfStart when both are provided.
336+
ocr: Gates all OCR. When true, PDFs get embedded-image OCR (recognized text inserted
337+
at each image's position in page reading order, preserving the text layer;
338+
pdf.start/pdf.end limit the page range), scanned PDFs with no text layer get
339+
full-document OCR, and raster images get their visible text transcribed. When
340+
false, no OCR runs: scanned PDFs may yield no content and images return only
341+
format/dimension metadata. Calls where OCR actually runs cost 5 credits instead
342+
of 1.
212343
213-
pdf_start: First 1-based PDF page to parse. When omitted, parsing starts at the first page.
344+
pdf: PDF page-range controls. Use start/end to limit parsing (and OCR when ocr=true)
345+
to an inclusive 1-based page range.
214346
215347
shorten_base64_images: Shorten base64-encoded image data in the Markdown output
216348
@@ -235,14 +367,11 @@ async def handle(
235367
timeout=timeout,
236368
query=await async_maybe_transform(
237369
{
238-
"base_url": base_url,
239370
"extension": extension,
240-
"filename": filename,
241371
"include_images": include_images,
242372
"include_links": include_links,
243373
"ocr": ocr,
244-
"pdf_end": pdf_end,
245-
"pdf_start": pdf_start,
374+
"pdf": pdf,
246375
"shorten_base64_images": shorten_base64_images,
247376
"use_main_content_only": use_main_content_only,
248377
},

0 commit comments

Comments
 (0)