filens-ocr-linux-x64 0.1.0 → 0.2.0

package/README.md

@@ -1,17 +1,19 @@
 # filens Agent CLI
 
 Read local documents and provide text and cited evidence for agents. Supports
-Chinese/English reading intents, plain text and JSON. Semantic reasoning is
+Chinese/English reading intents, text, JSON, Markdown, HTML, YAML, XML, CSV and
+TSV. Source layouts include PDF/OCR coordinates and logical document structure.
+Semantic reasoning is
 performed by the calling agent. All document parsing uses the shared WASM core.
 
 ```sh
 # Full: native RapidOCR NCNN and models included; no Python
-npm install -g filens
+npm install -g filens-cli
 filens '读取 "report.pdf"，以 JSON 输出'
 filens 'what can you do as json'
 
 # Lite: CLI + WASM; add native OCR when needed
-npm install -g filens-lite
+npm install -g filens-cli-lite
 filens setup-ocr
 filens 'read "scan.pdf" as json'
 ```
@@ -25,21 +27,39 @@ filens '帮我理解“会议纪要.docx”，用 JSON 返回'
 filens 'Please extract all text from "report.pdf" in JSON'
 filens '识别 "invoice.png" 中的文字，以 JSON 输出'
 filens '理解 "embedded-images.docx"，以 JSON 输出'
+filens 'read "report.docx" as markdown'
+filens '获取 "scan.pdf" 的布局，以 YAML 输出'
+filens layout invoice.png --format json
+filens read data.xlsx --format csv
 echo '{"operation":"understand","input":"report.pdf"}' | filens
 ```
 
 Natural language recognizes a bounded set of reading intents. `understand`
 returns source text and evidence for the calling agent; it does not generate
-an answer. One local file per request, text/JSON output. Summarization,
+an answer. One local file per request. Summarization,
 translation, questions, field extraction, page selection, URL fetching and file
-conversion are not implemented by this CLI. Some unsupported phrasings fall
+conversion to Office/PDF are not implemented by this CLI. Some unsupported phrasings fall
 back to reading the whole file: “only the first page” still reads all pages,
-“only list amounts” still returns all text, and “as yaml” falls back to text.
+“only list amounts” still returns all text.
 Use the JSON request schema or explicit `--format` when strict validation is
 required. Quote file paths inside natural-language sentences.
 
+JSON/YAML/XML preserve the full response protocol. Markdown/HTML export document
+content with headings and tables; `--layout` adds a metadata appendix. CSV/TSV
+export a single table as rows, or layout blocks as records when `--layout` is
+requested. `layout` defaults to JSON. PDF coordinates use points and a top-left
+origin; image OCR uses orientation-corrected pixels. Embedded-image coordinates
+are local to each image. Flow documents expose logical structure, without
+invented page positions. Detection scores do not measure transcription accuracy.
+
 OCR output may contain errors, including stray text detected in landscape
 photos. `ok=true` means extraction completed; it does not measure OCR accuracy.
+
+Supported PDF ToUnicode errors are repaired per font in the shared WASM core;
+unusable text mappings can require OCR even with a nonempty text layer. Missing
+fonts and incorrect rendering may still make numeric tables unreliable.
+JSON responses contain local source paths and document text; redact them before
+sharing. Private evaluation inputs and raw reports are excluded from packages.
 EXIF orientation is applied, while a physically inverted image may still have
 its text lines returned in reversed order.
 
@@ -61,4 +81,4 @@ one profile at a time. Platform packages are implementation components.
 [Protocol, limitations and source build](https://github.com/IceWhaleTech/filens/blob/main/docs/agent-cli.md)
 
 Locally built release tarballs can also be installed with
-`npm install -g /path/to/filens-0.1.0.tgz /path/to/filens-cli-PLATFORM-0.1.0.tgz /path/to/filens-ocr-PLATFORM-0.1.0.tgz`.
+`npm install -g /path/to/filens-cli-0.2.0.tgz /path/to/filens-cli-PLATFORM-0.2.0.tgz /path/to/filens-ocr-PLATFORM-0.2.0.tgz`.

package/THIRD-PARTY-NOTICES.md

@@ -20,8 +20,19 @@
 
 The Go host and native wrapper also contain Go, purego, wazero, imaging,
 gotokenizers, regexp2, protobuf and Go x/image/x/text components. Their notices
-are included in the package license directory. macOS wrapper builds use the
-pinned published source snapshot with deployment target 12.0.
+are included in the package license directory. All wrappers are rebuilt from the
+pinned published source snapshot with Go -trimpath. The macOS target is 12.0;
+Linux wrappers use Zig's glibc 2.17 target. wrapper-build.json records the
+source and rebuilt artifact hashes without workstation paths.
+
+The shared text WASM core uses lopdf 0.46.0 (MIT), ttf-parser 0.25.1
+(MIT OR Apache-2.0), and unicode-normalization 0.1.26
+((MIT OR Apache-2.0) AND Unicode-3.0) for PDF font mapping recovery.
+Their notices and those of the added transitive Rust components are included
+in the package license directory as pdf-unicode-rust.txt.
+
+Agent response YAML serialization uses go.yaml.in/yaml/v3 v3.0.4, with MIT
+and Apache-2.0 components. Its notice is included as yaml-v3-LICENSE.txt.
 
 This notice describes third-party components and does not grant a new license
 for filens or the Photos application code.