i18n(ko): fix hybrid X-to-Y terminology and update glossary overrides#1195
Open
lin-bot23 wants to merge 25 commits into
Open
i18n(ko): fix hybrid X-to-Y terminology and update glossary overrides#1195lin-bot23 wants to merge 25 commits into
lin-bot23 wants to merge 25 commits into
Conversation
|
Warning Review limit reached
Next review available in: 9 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Plus Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughThis PR standardizes Korean "X-to-Y" hybrid terminology (e.g., "텍스트-to-이미지") into natural Korean phrasing across tutorial and partner-node docs, refreshes MCP agent-tools wording, expands the glossary and docs.json navigation labels, adds a large V3 migration guide section, rewrites the startup-flags reference, and updates JS API example strings and translation metadata hashes throughout. ChangesMCP agent-tools wording
Korean X-to-Y terminology standardization
Korean documentation refresh and new content
Possibly related PRs
Time to wrap up this pun-tastic PR review — no loose threads left untied, and the docs are looking mighty fine, not just workflow-wined! 🐇 🚥 Pre-merge checks | ✅ 2✅ Passed checks (2 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
✨ Simplify code
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Restored files: - ko/custom-nodes/js/*.mdx (5 files) - Korean labels replaced with English - ko/development/cloud/overview.mdx - Korean comments replaced with English - ko/development/core-concepts/links.mdx - Korean type names replaced with English - ko/installation/comfyui_portable_windows.mdx - Korean dir tree replaced with English - ko/installation/update_comfyui.mdx - Korean tab titles replaced with English - ja/agent-tools/index.mdx - non-ko changes - zh/agent-tools/cloud.mdx - non-ko changes - zh/agent-tools/index.mdx - non-ko changes
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 28
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
ko/tutorials/video/cosmos/cosmos-predict2-video2world.mdx (1)
28-96: 🎯 Functional Correctness | 🔴 Critical | ⚡ Quick winEntire workflow section accidentally hidden inside an MDX comment.
Line 28 opens
{/*right after the intro<Card>, and the comment isn't closed until line 96 (*/}). Every following section —<UpdateReminder/>, the whole "Cosmos Predict2 Video2World 워크플로우" walkthrough, model download, and step-by-step instructions — is now commented out and will not render on the page at all. This turns the tutorial's core content invisible. Comment blocks don't rhyme well with missing content!Please remove the stray
{/*(and matching*/}) so the workflow section renders again.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@ko/tutorials/video/cosmos/cosmos-predict2-video2world.mdx` around lines 28 - 96, The entire Cosmos Predict2 Video2World tutorial content is unintentionally wrapped in an MDX comment, so it never renders. Remove the stray JSX comment wrapper around the workflow section in this MDX block (the opening at the start of the section and the matching closing near the end) so the `<UpdateReminder/>`, workflow steps, downloads, and instructions display again.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@ko/custom-nodes/js/javascript_overview.mdx`:
- Line 2: The document title in javascript_overview.mdx uses an inconsistent
brand spelling for JavaScript. Update the frontmatter title to use the standard
“JavaScript” capitalization consistently, and keep the wording aligned with the
upstream docs unless there is a deliberate localization-specific reason to
deviate.
In `@ko/custom-nodes/v3_migration.mdx`:
- Line 236: The Korean translation has in-page links that still point to English
fragment IDs, so the localized headings need stable explicit anchors. Update the
translated headings in the v3_migration.mdx content by adding matching IDs for
the sections referenced by [히든 입력](`#hidden-inputs`) and [지연
평가](`#lazy-evaluation-v1--v3`), using the heading text locations to keep the
anchor targets aligned even if the translation changes.
In `@ko/development/cloud/overview.mdx`:
- Around line 324-330: The page uses both “워크플로” and “워크플로우” for the same
concept, so standardize the wording across this document. Update the remaining
prose in the overview content that still says “워크플로우” to match the terminology
already used in the endpoint table and nearby “다음 단계” bullets, using the
relevant markdown sections in the overview document as the places to fix.
- Line 31: Align the tier terminology in overview.mdx so the same names are used
consistently across the page. Update the Korean prose in the subscription/API
access section and the tier table in the section referenced by the review so
they both use either the Korean tier names or the English tier names, matching
the page’s chosen locale style. Use the existing tier labels in the prose and
table content as the places to edit, especially the subscription sentence and
the tier table entries.
In `@ko/development/comfyui-server/api-examples.mdx`:
- Line 109: The docstring/comment text for the websockets API example uses an em
dash, which violates the markdown style rule for this MDX doc. Update the
affected English text in the `websockets_api_example.py` docstring/comment so it
uses a period, comma, colon, parentheses, or a second sentence instead of `—`,
and apply the same fix to the other matching comment in the same example
section.
- Around line 160-161: The preview-image comment around the `break` statement in
the example methods is inconsistent across the file, with one occurrence in
English and the matching `Method 3` comment still in Korean. Update the `Method
3` comment to match the same English wording used near the first `break` so the
duplicated preview-frame explanation is consistent throughout
`api-examples.mdx`.
In `@ko/development/comfyui-server/startup-flags.mdx`:
- Around line 33-46: The markdown blocks in startup-flags.mdx are left open,
causing the following headings and tables to render as code. Add the missing
closing ``` fences after each affected bash snippet in the sections around the
startup examples and flag tables so each code block is properly terminated
before the next heading or table. Use the surrounding section titles and the
repeated bash example blocks to locate and close each fence consistently.
In `@ko/development/core-concepts/links.mdx`:
- Around line 41-55: The data type labels in links.mdx are inconsistent with the
corresponding table in nodes.mdx, so align the terminology across both pages to
one localization standard. Update the links.mdx table entries in the data type
section to match the labels used in nodes.mdx (or vice versa) for the shared
symbols like Diffusion/확산 모델, Conditioning/컨디셔닝, Latent Image/잠재 이미지, Mask/마스크,
Number/숫자, and Mesh/메쉬. Keep the color mapping unchanged and ensure the same
terms are used consistently in both files.
In `@ko/installation/update_comfyui.mdx`:
- Line 211: The in-page anchor target for the troubleshooting section is
mismatched, so the link from the update guide will not resolve. Update the
heading text in the section containing 핵심 의존성 업데이트 문제 해결 and the corresponding
markdown link that points to the dependency update troubleshooting anchor so
they generate the same slug; verify the link and heading stay in sync for the
relevant section in update_comfyui.mdx.
In `@ko/specs/nodedef_json_1_0.mdx`:
- Line 1319: The final JSON example in nodedef_json_1_0.mdx is missing its
closing markdown code fence, which leaves the last `}` inside an open code
block. Update the affected fenced block near the end of the document by adding
the closing ``` delimiter after the JSON content so the Markdown/MDX parser can
terminate the block correctly. Use the surrounding example section in
nodedef_json_1_0.mdx to locate the unfinished fence.
In `@ko/specs/nodedef_json.mdx`:
- Line 538: The JSON example at the end of the MDX document is missing its
closing markdown code fence, so restore the terminating fence after the final
`}` in the `nodedef_json` example. Verify the fenced block around the JSON
sample remains properly balanced in the surrounding MDX content so renderers and
fence-aware tooling can parse it correctly.
In `@ko/specs/workflow_json_0.4.mdx`:
- Line 654: The JSON schema block in workflow_json_0.4.mdx is missing its
closing fence after the final brace, so the rest of the document is rendered as
code. Add the trailing closing ``` immediately after the final } to properly
terminate the fenced block and restore normal Markdown rendering.
In `@ko/tutorials/audio/ace-step/ace-step-v1.mdx`:
- Around line 53-55: The comparison phrase in the ACE-Step tutorial uses an
unrelated image-generation term, so update the introductory paragraph in
ace-step-v1.mdx to remove or replace the “이미지 기반 이미지 생성 워크플로우” reference with an
audio-relevant comparison. Use the existing audio tutorial context around the
ACE-Step workflow and Ksampler denoise explanation to align the wording with the
earlier text-to-audio section instead of copying the image tutorial language.
In `@ko/tutorials/controlnet/depth-t2i-adapter.mdx`:
- Line 50: The section heading in depth-t2i-adapter.mdx uses the transliterated
term “에셋,” which is inconsistent with the localized glossary used elsewhere.
Update this heading and the related workflow-asset wording to the agreed Korean
term, likely “자산,” and check nearby tutorial sections for the same terminology
so the assets-related labels are consistent across the cohort.
In `@ko/tutorials/flux/flux-2-klein.mdx`:
- Around line 24-35: Update the heading in the Flux.2 Klein tutorial to match
the document set’s Korean terminology by changing the title text in the affected
Card section from “워크플로” to “워크플로우”; keep the rest of the card copy unchanged
and use the same wording convention as the sibling tutorial files that already
reference “워크플로우”.
In `@ko/tutorials/flux/flux1-krea-dev.mdx`:
- Line 45: The tutorial label uses a Korean term for diffusion model that should
match the glossary choice used across the tutorial set. Update the term in this
document’s label and align the same wording in the other flux tutorial labels so
`확산 모델` and `디퓨전 모델` are not mixed; use the shared tutorial label text as the
reference point when making the terminology consistent.
In `@ko/tutorials/image/anima/anima.mdx`:
- Line 3: Normalize the terminology in anima.mdx so it matches the cohort’s
standard phrasing used in the other tutorial files; replace all ad-hoc variants
in the document (including the occurrences in the frontmatter description and
the body text around the affected sections) with one consistent Korean form like
“텍스트 기반 이미지 생성” or the agreed localized equivalent. Check the repeated
model-description wording in the page content and ensure all references in the
tutorial converge on the same phrase instead of mixing “텍스트-이미지”, “텍스트 → 이미지”,
and “애니메이션 텍스트-이미지 생성”.
In `@ko/tutorials/image/cosmos/cosmos-predict2-t2i.mdx`:
- Line 14: In the Cosmos-Predict2 tutorial sentence, fix the Korean particle
mismatch after the noun-ending “생성” so the phrasing is grammatically correct.
Update the text in the relevant markdown content to use “텍스트 기반 이미지
생성(Text2Image)과” and “비디오 기반 월드 생성(Video2World)을” instead of the current 와/를
forms, keeping the rest of the sentence unchanged.
In `@ko/tutorials/image/hidream/hidream-e1.mdx`:
- Line 139: Step 2 in the Hidream E1 tutorial is truncated and leaves invalid
markdown, so restore the missing QuadrupleCLIPLoader text-encoder verification
content in the same place where the step currently shows the broken “2. `”
fragment. Use the equivalent E1.1 section as the source for the full step text
and ensure the restored step includes the full 4-file list and closes the
markdown cleanly.
In `@ko/tutorials/image/omnigen/omnigen2.mdx`:
- Line 79: Correct the node name casing in the tutorial text: replace
CLipTextEncode with CLIPTextEncode in the prompt instructions. Update the
matching wording in the Omnigen tutorial content so the canonical node name is
used consistently wherever this step is described.
In `@ko/tutorials/partner-nodes/hunyuan3d/model-generation.mdx`:
- Around line 17-19: The terminology in the Hunyuan3D model generation page is
inconsistent with the Korean glossary and nearby partner-nodes docs. Update the
text in the model generation list and the related section in the same document
to use the glossary-consistent phrasing from the existing partner-nodes wording,
keeping the term aligned in both places within model-generation.mdx.
In `@ko/tutorials/partner-nodes/openai/dall-e-2.mdx`:
- Line 19: The Korean text in the DALL·E 2 tutorial contains a duplicated “생성”
phrase, so update the translated heading/content in the affected markdown to
read naturally as “텍스트 기반 이미지 생성”. Check the surrounding localized copy in the
same document for glossary or post-processing overlap, and make sure any
translation pass or term replacement logic that produced this text does not
append “생성” twice.
In `@ko/tutorials/partner-nodes/openrouter/llm.mdx`:
- Line 77: The phrasing in the tutorial uses an unnatural “텍스트-이미지” compound, so
update the sentence in llm.mdx to a more natural Korean expression that matches
the intended meaning, such as “이미지 생성 프롬프트” or “텍스트 기반 이미지 생성 프롬프트,” while
keeping the rest of the explanation intact. If similar “한글-한글” hyphenated
wording appears elsewhere in this content set, run the same localization/style
check to keep terminology consistent.
In `@ko/tutorials/partner-nodes/recraft/recraft-v4.mdx`:
- Around line 2-3: The frontmatter in the Recraft V4 tutorial has regressed from
Korean to English, so update the title and description in recraft-v4.mdx to
Korean. Use the existing localized markdown convention for ko/tutorials and make
sure the frontmatter matches the Korean audience; if the previous Korean text is
unavailable, translate the current title/description in place while keeping the
document’s meaning aligned with the content.
- Line 27: The Korean copy in this tutorial still uses hyphenated loan-style
phrases like “텍스트-이미지” and “텍스트-벡터,” which should be normalized across the
document. Update the affected content in this Recraft V4 tutorial to a
consistent natural Korean form such as “텍스트 기반 이미지 생성” and “텍스트 기반 벡터 생성,” and
make sure every occurrence in the same document is replaced uniformly so the
terminology stays consistent.
In `@ko/tutorials/partner-nodes/stability-ai/stable-audio.mdx`:
- Line 23: Fix the Korean particle agreement in the Stable Audio tutorial text:
the phrase in the `stable-audio.mdx` content currently uses “생성를” after “텍스트 기반
오디오 생성”, which is grammatically incorrect. Update the sentence to use the
correct object particle form “생성을 위해서는” in the affected markdown content,
keeping the rest of the wording unchanged.
In `@ko/tutorials/video/cosmos/cosmos-predict2-video2world.mdx`:
- Around line 14-27: The Cosmo-Predict2 tutorial still uses the mixed
“텍스트-이미지(Text2Image)” and “텍스트-이미지” wording, which should be aligned with the
cohort’s standardized localized term. Update the opening description and the
“텍스트-이미지 섹션은 다음을 참고하세요” reference in this document to use the glossary-driven
phrase “텍스트 기반 이미지 생성,” and make the Card title/body consistent with that
wording throughout the file.
In `@ko/tutorials/video/wan/wan2_2.mdx`:
- Line 96: Make the “browse templates” navigation wording consistent across this
document by updating all occurrences to the current menu label used by the
UI/docs.json layer. In the affected tutorial sections, align the path shown in
the instructions with the same phrasing everywhere, using the same symbols for
the menu trail in the workflow steps so readers don’t see mixed labels like “템플릿
탐색,” “템플릿 둘러보기,” and the English “Browse Templates” on the same page.
---
Outside diff comments:
In `@ko/tutorials/video/cosmos/cosmos-predict2-video2world.mdx`:
- Around line 28-96: The entire Cosmos Predict2 Video2World tutorial content is
unintentionally wrapped in an MDX comment, so it never renders. Remove the stray
JSX comment wrapper around the workflow section in this MDX block (the opening
at the start of the section and the matching closing near the end) so the
`<UpdateReminder/>`, workflow steps, downloads, and instructions display again.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 6d18d989-a111-4b03-93f5-cd95a21dcb0f
📒 Files selected for processing (99)
.github/scripts/i18n/glossary/overrides/ko.jsondocs.jsonja/agent-tools/index.mdxko/agent-tools/cloud.mdxko/agent-tools/partner-mcp.mdxko/changelog/index.mdxko/custom-nodes/js/javascript_about_panel_badges.mdxko/custom-nodes/js/javascript_bottom_panel_tabs.mdxko/custom-nodes/js/javascript_commands_keybindings.mdxko/custom-nodes/js/javascript_dialog.mdxko/custom-nodes/js/javascript_examples.mdxko/custom-nodes/js/javascript_overview.mdxko/custom-nodes/js/javascript_settings.mdxko/custom-nodes/js/javascript_sidebar_tabs.mdxko/custom-nodes/js/javascript_toast.mdxko/custom-nodes/js/subgraphs.mdxko/custom-nodes/overview.mdxko/custom-nodes/v3_migration.mdxko/custom-nodes/workflow_templates.mdxko/development/api-development/getting-an-api-key.mdxko/development/api-development/overview.mdxko/development/api-development/workflow-api-format.mdxko/development/cloud/openapi.mdxko/development/cloud/overview.mdxko/development/comfyui-server/api-examples.mdxko/development/comfyui-server/comms_messages.mdxko/development/comfyui-server/comms_overview.mdxko/development/comfyui-server/comms_routes.mdxko/development/comfyui-server/startup-flags.mdxko/development/core-concepts/custom-nodes.mdxko/development/core-concepts/dependencies.mdxko/development/core-concepts/links.mdxko/development/core-concepts/nodes.mdxko/development/core-concepts/properties.mdxko/installation/comfyui_portable_windows.mdxko/installation/desktop/faq.mdxko/installation/desktop/linux.mdxko/installation/desktop/overview.mdxko/installation/desktop/usage/migrate.mdxko/installation/desktop/usage/overview.mdxko/installation/desktop/usage/snapshots.mdxko/installation/install_custom_node.mdxko/installation/update_comfyui.mdxko/manager/pack-management.mdxko/manager/troubleshooting.mdxko/specs/nodedef_json.mdxko/specs/nodedef_json_1_0.mdxko/specs/workflow_json.mdxko/specs/workflow_json_0.4.mdxko/troubleshooting/model-issues.mdxko/tutorials/audio/ace-step/ace-step-v1.mdxko/tutorials/audio/stable-audio/stable-audio-1.mdxko/tutorials/basic/image-to-image.mdxko/tutorials/basic/upscale.mdxko/tutorials/controlnet/depth-t2i-adapter.mdxko/tutorials/flux/flux-2-klein.mdxko/tutorials/flux/flux1-krea-dev.mdxko/tutorials/image/anima/anima.mdxko/tutorials/image/cosmos/cosmos-predict2-t2i.mdxko/tutorials/image/ernie-image/ernie-image.mdxko/tutorials/image/hidream/hidream-e1.mdxko/tutorials/image/hidream/hidream-i1.mdxko/tutorials/image/hidream/hidream-o1.mdxko/tutorials/image/lens/lens.mdxko/tutorials/image/newbie-image/newbie-image-exp-0-1.mdxko/tutorials/image/omnigen/omnigen2.mdxko/tutorials/image/ovis/ovis-image.mdxko/tutorials/image/pixeldit/pixeldit.mdxko/tutorials/image/qwen/qwen-image-2512.mdxko/tutorials/image/z-image/z-image.mdxko/tutorials/partner-nodes/black-forest-labs/flux-1-1-pro-ultra-image.mdxko/tutorials/partner-nodes/bytedance/seedream-5-lite.mdxko/tutorials/partner-nodes/google/gemini.mdxko/tutorials/partner-nodes/hunyuan3d/hunyuan3d-3-0.mdxko/tutorials/partner-nodes/hunyuan3d/model-generation.mdxko/tutorials/partner-nodes/ideogram/ideogram-v3.mdxko/tutorials/partner-nodes/luma/luma-image-to-image.mdxko/tutorials/partner-nodes/luma/luma-text-to-image.mdxko/tutorials/partner-nodes/meshy/meshy-6.mdxko/tutorials/partner-nodes/moonvalley/moonvalley-video-generation.mdxko/tutorials/partner-nodes/openai/chat.mdxko/tutorials/partner-nodes/openai/dall-e-2.mdxko/tutorials/partner-nodes/openai/gpt-image-2.mdxko/tutorials/partner-nodes/openrouter/llm.mdxko/tutorials/partner-nodes/recraft/recraft-v4.mdxko/tutorials/partner-nodes/reve/reve-image.mdxko/tutorials/partner-nodes/rodin/model-generation.mdxko/tutorials/partner-nodes/runway/image-generation.mdxko/tutorials/partner-nodes/stability-ai/stable-audio.mdxko/tutorials/partner-nodes/stability-ai/stable-image-ultra.mdxko/tutorials/partner-nodes/tripo/tripo-3-1.mdxko/tutorials/video/cosmos/cosmos-predict2-video2world.mdxko/tutorials/video/hunyuan/hunyuan-video-1-5.mdxko/tutorials/video/wan/fun-control.mdxko/tutorials/video/wan/vace.mdxko/tutorials/video/wan/wan-alpha.mdxko/tutorials/video/wan/wan2_2.mdxzh/agent-tools/cloud.mdxzh/agent-tools/index.mdx
💤 Files with no reviewable changes (1)
- ko/installation/desktop/linux.mdx
- Fix Korean grammar particles (와→과, 를→을) - Fix CLIPTextEncode casing (CLipTextEncode→CLIPTextEncode) - Remove duplicate '생성 생성' → '생성' - Unify '다중뷰' → '멀티뷰' - Add explicit heading anchors for in-page links - Replace em dashes with colons/semicolons - Close unclosed code fences in startup-flags - Normalize wan2_2 menu paths to unified Korean - Replace hyphenated terms (텍스트-이미지→텍스트 기반 이미지 생성) - Restore Korean frontmatter in recraft-v4
comfyui-wiki
reviewed
Jul 1, 2026
|
|
||
| // 모든 토스트 제거 | ||
| app.extensionManager.toast.removeAll(); | ||
| ``` |
Member
There was a problem hiding this comment.
You shouldn't remove this line
Contributor
Author
There was a problem hiding this comment.
✅ Fixed in commit a1a6473 - restored the missing closing ``` fence at end of file.
Contributor
Author
There was a problem hiding this comment.
✅ Fixed in commit a1a6473 - restored the missing closing ``` fence.
|
|
||
| ```python | ||
| """websockets_api_example.py — WebSocket을 통해 실행 상태를 모니터링하고, /history를 통해 다운로드합니다.""" | ||
| """websockets_api_example.py — Monitor execution via WebSocket, download via /history.""" |
Member
There was a problem hiding this comment.
I think it's better to keep it in the Korean translation, not English
Contributor
Author
There was a problem hiding this comment.
✅ Fixed in commit a1a6473 - restored Korean comments/docstrings from upstream/main. Only the em dashes were kept fixed (— → :/;) per the CodeRabbit suggestion.
| break # 실행 완료 | ||
| # 이진 프레임은 미리보기 이미지이므로 여기서 건너뜁니다 | ||
| break # Execution done | ||
| # Binary frames are preview images — skip them here |
Member
There was a problem hiding this comment.
Please translate it into Korean
| print(f"Got {len(images)} output node(s) with images.") | ||
|
|
||
| # 이미지 표시(필요한 경우 Pillow 사용): | ||
| # Display the images (requires Pillow): |
Member
There was a problem hiding this comment.
There is also a translation issue.
| body.append('node_id', node_id); | ||
| api.fetchApi("/my_new_path", { method: "POST", body, }); | ||
| } | ||
| ``` |
Member
There was a problem hiding this comment.
Please don't remove this line
| }, | ||
| "$schema": "http://json-schema.org/draft-07/schema#" | ||
| } | ||
| ``` |
Member
There was a problem hiding this comment.
Please don't remove this line
| │ │ └── ovis_image_bf16.safetensors | ||
| │ └── 📂 vae/ | ||
| │ └── ae.safetensors | ||
| ``` |
Member
There was a problem hiding this comment.
Please don't remove this line
| │ │ └── qwen_image_2512_fp8_e4m3fn.safetensors | ||
| │ └── 📂 vae/ | ||
| │ └── qwen_image_vae.safetensors | ||
| ``` |
Member
There was a problem hiding this comment.
Please don't remove this line
Comment on lines
-14
to
-16
| ## Reve 소개 | ||
|
|
||
| **주요 기능:** |
Member
There was a problem hiding this comment.
Why has this section been removed?
Contributor
Author
There was a problem hiding this comment.
✅ Fixed in commit 9021839 - restored the 'Reve 소개' section from upstream/main.
| defaultValue: false, | ||
| tooltip: "이것은 도움이 되는 정보입니다", | ||
| } | ||
| ``` No newline at end of file |
Member
There was a problem hiding this comment.
Please don't remove this line
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
ko/tutorials/image/omnigen/omnigen2.mdx (1)
94-94: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick winAlt text mismatched with section content.
Line 94 sits under "이미지 편집 워크플로우" (image editing) and links
image_omnigen2_image_edit.png, but the alt text reads "텍스트 기반 이미지 생성 워크플로우" (text-based image generation workflow) — same as the unrelated t2i screenshot's alt text at line 66. This mismatch predates this PR (line-range details show it was already generic/wrong before, just term-substituted here), but since this cohort is touching this exact line, it's worth correcting now rather than perpetuating it.✏️ Suggested alt-text fix
- +🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@ko/tutorials/image/omnigen/omnigen2.mdx` at line 94, Update the image alt text under the “이미지 편집 워크플로우” section so it matches the linked `image_omnigen2_image_edit.png` content; the current alt text in `omnigen2.mdx` still describes the text-based generation workflow. Use the nearby markdown image entry for the edit workflow to rename the alt text to an image-editing description, keeping it distinct from the separate t2i screenshot’s label.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@ko/tutorials/partner-nodes/hunyuan3d/model-generation.mdx`:
- Line 19: The glossary term for multiview-to-model is inconsistent with the
tutorial heading and bullet text. Update the title and any matching bullet in
the Hunyuan3D markdown so they both use “멀티뷰 기반 3D 모델 생성” consistently, keeping
the existing multiview-to-model terminology aligned across the document.
---
Outside diff comments:
In `@ko/tutorials/image/omnigen/omnigen2.mdx`:
- Line 94: Update the image alt text under the “이미지 편집 워크플로우” section so it
matches the linked `image_omnigen2_image_edit.png` content; the current alt text
in `omnigen2.mdx` still describes the text-based generation workflow. Use the
nearby markdown image entry for the edit workflow to rename the alt text to an
image-editing description, keeping it distinct from the separate t2i
screenshot’s label.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: ASSERTIVE
Plan: Pro Plus
Run ID: 7dae7c0b-bbf7-45ae-bc9a-b4334978db3a
📒 Files selected for processing (13)
ko/custom-nodes/v3_migration.mdxko/development/comfyui-server/api-examples.mdxko/development/comfyui-server/startup-flags.mdxko/tutorials/image/anima/anima.mdxko/tutorials/image/cosmos/cosmos-predict2-t2i.mdxko/tutorials/image/omnigen/omnigen2.mdxko/tutorials/partner-nodes/hunyuan3d/model-generation.mdxko/tutorials/partner-nodes/openai/dall-e-2.mdxko/tutorials/partner-nodes/openrouter/llm.mdxko/tutorials/partner-nodes/recraft/recraft-v4.mdxko/tutorials/partner-nodes/stability-ai/stable-audio.mdxko/tutorials/video/cosmos/cosmos-predict2-video2world.mdxko/tutorials/video/wan/wan2_2.mdx
| - 다중뷰-to-3D 생성 | ||
| - 텍스트 기반 3D 생성 | ||
| - 이미지 기반 3D 생성 | ||
| - 멀티뷰 기반 3D 생성 |
There was a problem hiding this comment.
🎯 Functional Correctness | 🔵 Trivial | 💤 Low value
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Verify if 'multiview-to-3d' or similar exists in glossary, or if '모델' should be added
rg -n "multiview" .github/scripts/i18n/glossary/overrides/ko.jsonRepository: Comfy-Org/docs
Length of output: 200
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Show the relevant section of the Korean tutorial file with line numbers.
sed -n '1,110p' ko/tutorials/partner-nodes/hunyuan3d/model-generation.mdx | cat -n
# Show nearby glossary entries for comparison.
sed -n '1,80p' .github/scripts/i18n/glossary/overrides/ko.json | cat -nRepository: Comfy-Org/docs
Length of output: 5247
🏁 Script executed:
#!/bin/bash
set -euo pipefail
# Locate the English source file and show the matching sections.
src="tutorials/partner-nodes/hunyuan3d/model-generation.mdx"
if [ -f "$src" ]; then
sed -n '1,110p' "$src" | cat -n
else
echo "Missing source file: $src" >&2
exit 1
fi
# Check whether other locales use "모델" in the analogous headings/bullets.
rg -n "멀티뷰 기반 3D|text-to-3d|image-to-3d|multiview-to-model" ko tutorials zh jaRepository: Comfy-Org/docs
Length of output: 4357
"멀티뷰 기반 3D 생성"을 "멀티뷰 기반 3D 모델 생성"으로 맞춰 주세요. multiview-to-model glossary와 bullet/제목(19, 63행)이 어긋납니다.
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@ko/tutorials/partner-nodes/hunyuan3d/model-generation.mdx` at line 19, The
glossary term for multiview-to-model is inconsistent with the tutorial heading
and bullet text. Update the title and any matching bullet in the Hunyuan3D
markdown so they both use “멀티뷰 기반 3D 모델 생성” consistently, keeping the existing
multiview-to-model terminology aligned across the document.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fix hybrid English-Korean X-to-Y terms across Korean (ko/) documentation.
Changes
Glossary overrides (
.github/scripts/i18n/glossary/overrides/ko.json)Added 12 new preferred term mappings:
audio-to-audio,video-to-audio,text-to-speech,text-to-3d,image-to-3d,frame-interpolationtext to image,image to image,text to video,image to video,reference to image,reference to videoKorean documentation (
ko/)pnpm translate --force(batch 1)텍스트-to-image→텍스트 기반 이미지 생성)Summary
-to-+ English/Korean suffix) are replaced with pure Korean equivalents per the glossary