Adds the omniread wiki to the hub alongside lib and mcp and completes the library nav with csv/xlsx groups, picking up regenerated MCP modules.
41 lines
2.7 KiB
JSON
41 lines
2.7 KiB
JSON
{
|
|
"module": "omniread.pdf.client",
|
|
"content": {
|
|
"path": "omniread.pdf.client",
|
|
"docstring": "# Summary\n\nPDF client abstractions for OmniRead.\n\nThis module defines the **client layer** responsible for retrieving raw PDF\nbytes from a concrete backing store.\n\nClients provide low-level access to PDF binaries and are intentionally\ndecoupled from scraping and parsing logic. They do not perform validation,\ninterpretation, or content extraction.\n\nTypical backing stores include:\n\n- Local filesystems\n- Object storage (S3, GCS, etc.)\n- Network file systems",
|
|
"objects": {
|
|
"BasePDFClient": {
|
|
"name": "BasePDFClient",
|
|
"kind": "class",
|
|
"path": "omniread.pdf.client.BasePDFClient",
|
|
"signature": null,
|
|
"docstring": "Abstract client responsible for retrieving PDF bytes.\n\nRetrieves bytes from a specific backing store (filesystem, S3, FTP, etc.).\n\nNotes:\n **Responsibilities:**\n\n - Implementations must accept a source identifier appropriate to\n the backing store.\n - Return the full PDF binary payload.\n - Raise retrieval-specific errors on failure.",
|
|
"members": {
|
|
"fetch": {
|
|
"name": "fetch",
|
|
"kind": "function",
|
|
"path": "omniread.pdf.client.BasePDFClient.fetch",
|
|
"signature": "fetch(source: Any) -> bytes",
|
|
"docstring": "Fetch raw PDF bytes from the given source.\n\nArgs:\n source (Any):\n Identifier of the PDF location, such as a file path, object storage key, or remote reference.\n\nReturns:\n bytes:\n Raw PDF bytes.\n\nRaises:\n Exception:\n Retrieval-specific errors defined by the implementation."
|
|
}
|
|
}
|
|
},
|
|
"FileSystemPDFClient": {
|
|
"name": "FileSystemPDFClient",
|
|
"kind": "class",
|
|
"path": "omniread.pdf.client.FileSystemPDFClient",
|
|
"signature": null,
|
|
"docstring": "PDF client that reads from the local filesystem.\n\nNotes:\n **Guarantees:**\n\n - This client reads PDF files directly from the disk and returns\n their raw binary contents.",
|
|
"members": {
|
|
"fetch": {
|
|
"name": "fetch",
|
|
"kind": "function",
|
|
"path": "omniread.pdf.client.FileSystemPDFClient.fetch",
|
|
"signature": "fetch(path: Path) -> bytes",
|
|
"docstring": "Read a PDF file from the local filesystem.\n\nArgs:\n path (Path):\n Filesystem path to the PDF file.\n\nReturns:\n bytes:\n Raw PDF bytes.\n\nRaises:\n FileNotFoundError:\n If the path does not exist.\n ValueError:\n If the path exists but is not a file."
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
} |