@jarkkojs/readseek 0.7.9 → 0.8.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +16 -20
  2. package/package.json +5 -5
package/README.md CHANGED
@@ -53,7 +53,7 @@ tools to OpenCode. Add it to `opencode.json`:
53
53
 
54
54
  The bundled `readseek.vim` plugin provides anchored reads, structural navigation,
55
55
  parse diagnostics, search, references, and renames. It requires Vim9 with `+job`,
56
- `+channel`, `+popupwin`, and `+textprop`; Neovim is not supported.
56
+ `+channel`, `+timers`, `+popupwin`, and `+textprop`; Neovim is not supported.
57
57
 
58
58
  Install it with a Vim plugin manager, such as `Plug 'jarkkojs/readseek'`, then run
59
59
  `:ReadSeekInstall` to install the matching prebuilt binary. Downloads are explicit
@@ -93,38 +93,34 @@ printf '%s\n' 'fn main() {}' | readseek identify stdin:scratch.rs:1 --column 4
93
93
  ## Images and PDFs
94
94
 
95
95
  `detect` reports image metadata and PDF page counts. `read` returns bounded base64
96
- images by default; select local analysis with `--image`:
96
+ images by default; select local analysis with `--vision-mode`:
97
97
 
98
98
  ```sh
99
- readseek read photo.jpg # default: bounded base64 image
100
- readseek read photo.jpg --image caption # detailed natural-language caption
101
- readseek read photo.jpg --image objects # object labels + bounding boxes
102
- readseek read photo.jpg --image ocr # extracted text
103
- readseek read photo.jpg --image all # caption, objects, and OCR in one pass
99
+ readseek read photo.jpg # default: bounded base64 image
100
+ readseek read photo.jpg --vision-mode caption # detailed natural-language caption
101
+ readseek read photo.jpg --vision-mode objects # object labels + bounding boxes
102
+ readseek read photo.jpg --vision-mode ocr # extracted text
103
+ readseek read photo.jpg --vision-mode all # caption, objects, and OCR in one pass
104
104
  ```
105
105
 
106
- Vision analysis uses the `balanced` profile by default. Choose `fast` for lower
107
- latency or `accurate` for dense OCR and small objects:
106
+ Vision analysis uses the `low` level by default. Start at `low`, then increase to
107
+ `medium` or `high` only when additional detail is needed:
108
108
 
109
109
  ```sh
110
- readseek read photo.jpg --image caption --vision-profile fast
111
- readseek read scan.png --image ocr --vision-profile accurate
110
+ readseek read photo.jpg --vision-mode caption --vision-level low
111
+ readseek read scan.png --vision-mode ocr --vision-level high
112
112
  ```
113
113
 
114
- `--vision-diagnostics` writes cache status, devices, token counts, timings,
115
- throughput, and peak RSS as JSON to stderr. `--vision-benchmark N` runs one warmup
116
- and `N` measured iterations for an image file:
114
+ Set `RUST_LOG=tracing` to emit vision cache and inference traces on standard error.
117
115
 
118
- ```sh
119
- readseek read fixture.png --image all --vision-benchmark 5 \
120
- >result.json 2>benchmark.json
121
- ```
116
+ Run `cargo bench --features vision-bench --bench vision` for data-driven comparisons
117
+ defined in `benches/vision.txt`.
122
118
 
123
119
  Set `READSEEK_VISION_THREADS` to a positive integer to override Rayon's worker
124
120
  count. More workers may improve throughput at the cost of CPU and memory.
125
121
 
126
122
  PDF reads return page-tagged Markdown and embedded images. Use `--page` to select a
127
- page; `--image` applies to each embedded image. After `readseek init`, `view` creates
123
+ page; `--vision-mode` applies to each embedded image. After `readseek init`, `view` creates
128
124
  or reuses a structural PDF index that can be narrowed by page, node, kind, or depth.
129
125
  Line/hash suffixes, `--end`, `--limit`, and `--language` do not apply to visual files.
130
126
 
@@ -138,7 +134,7 @@ can be slow.
138
134
  Map-dependent commands update them on demand and find `.readseek/` by walking up
139
135
  from the target; `--readseek-dir` selects one explicitly. PDF indexes and extracted
140
136
  assets live in `.readseek/documents/`, while image analysis results are cached under
141
- `.readseek/vision/`. Image cache entries are profile-specific.
137
+ `.readseek/vision/`. Image cache entries are level-specific.
142
138
 
143
139
  ## Documentation
144
140
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@jarkkojs/readseek",
3
- "version": "0.7.9",
3
+ "version": "0.8.0",
4
4
  "description": "A structural read command",
5
5
  "license": "Apache-2.0 AND LGPL-2.1-or-later",
6
6
  "homepage": "https://github.com/jarkkojs/readseek#readme",
@@ -21,10 +21,10 @@
21
21
  "LICENSE-LGPL-2.1"
22
22
  ],
23
23
  "optionalDependencies": {
24
- "@jarkkojs/readseek-darwin-arm64": "0.7.9",
25
- "@jarkkojs/readseek-linux-arm64": "0.7.9",
26
- "@jarkkojs/readseek-linux-x64": "0.7.9",
27
- "@jarkkojs/readseek-win32-x64": "0.7.9"
24
+ "@jarkkojs/readseek-darwin-arm64": "0.8.0",
25
+ "@jarkkojs/readseek-linux-arm64": "0.8.0",
26
+ "@jarkkojs/readseek-linux-x64": "0.8.0",
27
+ "@jarkkojs/readseek-win32-x64": "0.8.0"
28
28
  },
29
29
  "publishConfig": {
30
30
  "access": "public"