claude2md 0.1__py3-none-any.whl

claude2md/__init__.py

@@ -0,0 +1,339 @@
+#!/usr/bin/env python3
+
+# SPDX-License-Identifier: CC0-1.0
+
+"""Convert Claude.ai or Claude Code chats to Markdown"""
+
+__version__ = "0.1"
+
+from contextlib import suppress
+from dataclasses import dataclass, field
+from enum import Enum
+from json import JSONDecodeError
+from uuid import UUID
+
+import argparse
+import json
+import os
+import re
+import sys
+
+NULL_UUID = "00000000-0000-4000-8000-000000000000"
+
+
+class Role(Enum):
+    USER = "user"
+    ASSISTANT = "assistant"
+
+    @classmethod
+    def from_json(cls, value):
+        match value:
+            case "user" | "human":
+                return cls.USER
+            case "assistant":
+                return cls.ASSISTANT
+            case _:
+                raise ValueError(f'Unknown role: "{value}"')
+
+
+@dataclass(slots=True, frozen=True)
+class Block:
+    type: str
+    content: str
+
+    @classmethod
+    def from_json(cls, obj):
+        match obj["type"]:
+            case "text":
+                return cls("text", obj["text"])
+            case "thinking":
+                return cls("thinking", obj["thinking"])
+            case "token_budget" | "tool_result" | "tool_use" | "voice_note":
+                return None
+            case block_type:
+                raise NotImplementedError(f'Unsupported block type "{block_type}"')
+
+
+@dataclass(slots=True, frozen=True)
+class Message:
+    uuid: str
+    parent: str | None
+    role: Role
+    blocks: list[Block]
+    attachments: list
+
+    @classmethod
+    def from_json(cls, data):
+        # Try to get content from different possible locations
+        content = data.get("message", {}).get("content") or data.get("content")
+        if not content:
+            raise KeyError("No content field found in message data")
+
+        # Handle content as either a string or an array of block objects
+        if isinstance(content, str):
+            # Simple string content - create a text block
+            blocks = [Block("text", content)]
+        elif isinstance(content, list):
+            # Array of block objects
+            blocks = [
+                block for block in (Block.from_json(obj) for obj in content) if block
+            ]
+        else:
+            blocks = []
+        parent = data.get("parentUuid") or data.get("parent_message_uuid")
+        # Treat null UUID as None (no parent)
+        if parent == NULL_UUID:
+            parent = None
+
+        # Try to get role from different possible locations
+        role_value = data.get("message", {}).get("role") or data.get("sender")
+        if not role_value:
+            raise KeyError("No role field found in message data")
+
+        return cls(
+            uuid=data["uuid"],
+            parent=parent,
+            role=Role.from_json(role_value),
+            blocks=blocks,
+            attachments=data.get("attachments", []),
+        )
+
+    def render(self, thinking=True):
+        parts = []
+
+        if self.role == Role.USER:
+            if self.attachments:
+                lines = ["<documents>"]
+                for index, attachment in enumerate(self.attachments, 1):
+                    file_name = attachment.get("file_name")
+                    file_type = attachment.get("file_type")
+                    extracted = attachment["extracted_content"]
+
+                    doc = f'<document index="{index}"'
+                    if file_type:
+                        doc += f' media_type="{file_type}"'
+                    doc += ">"
+                    if file_name:
+                        doc += f"<source>{file_name}</source>"
+                    doc += (
+                        f"<document_content>{extracted}</document_content></document>"
+                    )
+                    lines.append(doc)
+                lines.append("</documents>")
+                parts.append("\n".join(lines))
+
+            for block in self.blocks:
+                parts.append(block.content)
+
+            text = "\n\n".join(parts)
+            return re.sub(r"^", "> ", text, flags=re.MULTILINE)
+
+        elif self.role == Role.ASSISTANT:
+            for block in self.blocks:
+                if block.type == "thinking" and not thinking:
+                    continue
+                if block.type == "thinking":
+                    parts.append(f"<thinking>\n{block.content}\n</thinking>")
+                else:
+                    parts.append(block.content)
+
+            return "\n\n".join(parts)
+
+        return ""
+
+
+@dataclass(slots=True)
+class Chat:
+    file: object
+
+    messages: dict = field(init=False, default_factory=dict)
+    metadata: dict = field(init=False, default_factory=dict)
+    leaves: list = field(init=False, default_factory=list)
+
+    def __post_init__(self):
+        unparsed = self.file.read()
+
+        try:
+            parsed = json.loads(unparsed)
+            for msg_data in parsed["chat_messages"]:
+                msg = Message.from_json(msg_data)
+                self.messages[msg.uuid] = msg
+            self.metadata = parsed
+        except (JSONDecodeError, KeyError, TypeError):
+            for line in unparsed.split("\n"):
+                with suppress(JSONDecodeError, KeyError, TypeError):
+                    msg = Message.from_json(json.loads(line))
+                    self.messages[msg.uuid] = msg
+
+        all_parents = {msg.parent for msg in self.messages.values() if msg.parent}
+        self.leaves = [uuid for uuid in self.messages.keys() if uuid not in all_parents]
+
+    def print_branch(self, uuid, user=True, assistant=True, thinking=False, title=None):
+        msg = self.messages[uuid]
+
+        if msg.parent:
+            self.print_branch(msg.parent, user, assistant, thinking, title)
+        elif title:
+            chat_name = self.metadata.get("name")
+            if title is True:
+                title_text = chat_name or "Untitled"
+                print(f"# {title_text}\n")
+            elif chat_name:
+                print(f"# {chat_name}\n")
+
+        if msg.role == Role.USER and not user:
+            return
+        if msg.role == Role.ASSISTANT and not assistant:
+            return
+
+        rendered = msg.render(thinking=thinking)
+        if rendered:
+            print(rendered)
+            print()
+
+    def print_leaves(self, uuids=None, file=None):
+        if uuids is None:
+            uuids = self.leaves
+
+        width = 80
+        if sys.stdout.isatty():
+            with suppress(AttributeError, OSError):
+                width = os.get_terminal_size().columns
+        preview_len = max(20, width - 36 - 4)
+
+        for uuid in uuids:
+            current = uuid
+            preview = ""
+
+            while current:
+                msg = self.messages[current]
+                if msg.role == Role.USER:
+                    for block in msg.blocks:
+                        if block.type == "text":
+                            preview = block.content
+                            break
+                    if preview:
+                        break
+                current = msg.parent
+
+            if preview:
+                preview = preview.replace("\n", " ").replace("\r", " ")
+                preview = " ".join(preview.split())
+                if len(preview) > preview_len:
+                    preview = preview[: preview_len - 1] + "…"
+            else:
+                preview = ""
+
+            print(f"{uuid}\t{preview}", file=file)
+
+    def get_leaf(self, prefix=""):
+        with suppress(ValueError):
+            return str(UUID(prefix))
+
+        matches = [uuid for uuid in self.leaves if uuid.startswith(prefix)]
+
+        match matches:
+            case []:
+                print(f'Error: No leaf matches prefix "{prefix}"', file=sys.stderr)
+                return None
+            case [single]:
+                return single
+            case multiple:
+                print(
+                    f'Error: Multiple leaves match prefix "{prefix}":',
+                    file=sys.stderr,
+                )
+                self.print_leaves(multiple, file=sys.stderr)
+                return None
+
+
+def main():
+    sys.setrecursionlimit(100_000)  # lol
+
+    parser = argparse.ArgumentParser(
+        description="Convert Claude.ai or Claude Code chats to Markdown"
+    )
+    parser.add_argument(
+        "file",
+        nargs="?",
+        type=argparse.FileType("r"),
+        default=sys.stdin,
+        help="JSON file to convert (default: stdin)",
+    )
+
+    nav_group = parser.add_mutually_exclusive_group()
+    nav_group.add_argument(
+        "--branches",
+        dest="leaves",
+        action="store_true",
+        help="List all branch message UUIDs",
+    )
+    nav_group.add_argument(
+        "--branch", dest="leaf", metavar="UUID", help="Show chain to specific branch"
+    )
+
+    parser.add_argument(
+        "--user",
+        dest="user",
+        action="store_true",
+        default=True,
+        help="Show user messages (default)",
+    )
+    parser.add_argument(
+        "--no-user", dest="user", action="store_false", help="Hide user messages"
+    )
+    parser.add_argument(
+        "--human", dest="user", action="store_true", help=argparse.SUPPRESS
+    )
+    parser.add_argument(
+        "--no-human", dest="user", action="store_false", help=argparse.SUPPRESS
+    )
+    parser.add_argument(
+        "--assistant",
+        dest="assistant",
+        action="store_true",
+        default=True,
+        help="Show assistant (default)",
+    )
+    parser.add_argument(
+        "--no-assistant", dest="assistant", action="store_false", help="Hide assistant"
+    )
+    parser.add_argument(
+        "--thinking",
+        dest="thinking",
+        action="store_true",
+        default=False,
+        help="Show thinking blocks",
+    )
+    parser.add_argument(
+        "--no-thinking",
+        dest="thinking",
+        action="store_false",
+        help="Hide thinking (default)",
+    )
+    parser.add_argument(
+        "--title",
+        dest="title",
+        action="store_true",
+        default=None,
+        help='Always show title (or "Untitled" if none)',
+    )
+    parser.add_argument(
+        "--no-title", dest="title", action="store_false", help="Never show title"
+    )
+    args = parser.parse_args()
+
+    chat = Chat(args.file)
+
+    if args.leaves:
+        chat.print_leaves()
+    elif uuid := (
+        (args.leaf and chat.get_leaf(args.leaf))
+        or chat.metadata.get("current_leaf_message_uuid")
+        or chat.get_leaf()
+    ):
+        chat.print_branch(uuid, args.user, args.assistant, args.thinking, args.title)
+
+
+if __name__ == "__main__":
+    main()

claude2md/__main__.py

@@ -0,0 +1,4 @@
+from claude2md import main
+
+if __name__ == "__main__":
+    main()

claude2md-0.1.dist-info/METADATA

@@ -0,0 +1,120 @@
+Metadata-Version: 2.4
+Name: claude2md
+Version: 0.1
+Summary: Convert Claude.ai or Claude Code chats to Markdown
+Keywords: claude,markdown,chat,export,converter
+Author-email: Ultrathinkers Anonymous <ultrathink@twilligon.com>
+Requires-Python: >= 3.10
+Description-Content-Type: text/markdown
+License-Expression: CC0-1.0
+Classifier: Development Status :: 4 - Beta
+Classifier: Environment :: Console
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Text Processing :: Markup :: Markdown
+Classifier: Topic :: Utilities
+License-File: LICENSE
+Project-URL: Issues, https://github.com/twilliddgon/claude2md/issues
+Project-URL: Repository, https://github.com/twilligon/claude2md
+
+# `claude2md`
+
+Convert Claude.ai or Claude Code chat exports to Markdown format.
+
+`claude2md` converts chats exported from Claude.ai or Claude Code to Markdown format. This lets you efficiently show entire chats to other models or instances until you struggle to remember which transcript of a chat full of transcripts of chats full of transcripts of chats is the right one (aka slopception).
+
+## Features
+
+You can filter the exports to any subset of user, assistant, and thinking blocks. For context management I've found it's often helpful to only export user messages, though obviously for the strongest "continuity" between an exported chat transcript and its consumer you want all context from the prior chat. Though Claude typically does not get to see its past thinking, so `--no-thinking` (our default) ought to be just as good if not better than including thinking blocks (by virtue of being on-distribution).
+
+We try to include Claude.ai attachments when they're available in the export, but the format we parse right now does not have them embedded in the JSON or anything, so in practice only attachments with `extracted_content`, i.e. text files, markdown, etc. will be present in the transcripts.
+
+You can branch conversations on Claude.ai (and to a lesser extent in Claude Code), and `claude2md` can export any branch. You can list the branches of a chat and their UUIDs with `claude2md --branches` and export a specific branch with `claude2md --branch UUID`.
+
+We don't support "Claude Code on the Web" yet, though ostensibly if you "teleport" a Web session to a local Claude Code instance, you can export the history as normal from there.
+
+## Install
+
+The blazing fast and memory safe way:
+
+    $ uvx claude2md --help  # installed on demand
+
+The traditional way:
+
+    $ pip install claude2md
+    $ claude2md --help
+
+The bleeding-edge way:
+
+    $ git clone https://github.com/twilligon/claude2md
+    $ cd claude2md
+    $ python3 -m venv venv; . venv/bin/activate  # you probably want a venv
+    $ pip install -e .
+    $ claude2md --help
+
+## Usage
+
+    $ claude2md --help
+    usage: claude2md [-h] [--branches | --branch UUID] [--user] [--no-user]
+                     [--assistant] [--no-assistant] [--thinking]
+                     [--no-thinking] [--title] [--no-title]
+                     [file]
+
+    Convert Claude.ai or Claude Code chats to Markdown
+
+    positional arguments:
+      file            JSON file to convert (default: stdin)
+
+    options:
+      -h, --help      show this help message and exit
+      --branches      List all branch message UUIDs
+      --branch UUID   Show chain to specific branch
+      --user          Show user messages (default)
+      --no-user       Hide user messages
+      --assistant     Show assistant (default)
+      --no-assistant  Hide assistant
+      --thinking      Show thinking blocks
+      --no-thinking   Hide thinking (default)
+      --title         Always show title (or "Untitled" if none)
+      --no-title      Never show title
+
+## Examples
+
+#### Claude.ai
+
+    $ claude2md --title --thinking ~/.local/share/claude-backup/f50d0792-fa79-4d2d-a945-c8f709e9e7a7/c6a33008-39b3-45ed-a00e-413141c89a90.json | head -n20
+    # Python packaging complexity
+
+    > why is python packaging such a clusterfuck? ultrathink
+
+    <thinking>
+    The user is asking about Python packaging and why it's so complicated/messy. They've also said "ultrathink" which I interpret as wanting a very thorough, deep analysis.
+
+    This is a great question that many Python developers have strong feelings about. Let me think through the historical and technical reasons why Python packaging became so complex.
+
+    Key issues to cover:
+    1. Historical evolution - how we got here
+    2. Multiple competing tools and standards
+    3. The various problems each tool tries to solve
+    4. Why there's no single "right" way
+    5. The current state and attempts to fix things
+
+    Let me structure this thoughtfully and comprehensively since they asked for "ultrathink."
+    </thinking>
+
+    Python packaging *is* notoriously complex, and there are deep historical and technical reasons for this. Let me break down why it became such a mess:
+
+#### Claude Code
+
+    $ claude2md ~/.claude/projects/-home-claude-claude2md/0c8de429-fb29-4096-925e-6d81a0437f09.jsonl --branch 670e0906-013b-4c83-b7c7-ecf8e12aab3e | head -n3
+    > rename --leaves and --leaf to --branches and --branch. no need to change how we refer to this stuff internally, only user facing matters. in @claude2md/__init__.py . ultrathink
+
+    I'll help you rename the command-line arguments from `--leaves`/`--leaf` to `--branches`/`--branch` in the user-facing interface. Let me start by creating a todo list for this task.
+
+## License
+
+`claude2md` is dedicated to the public domain where possible via CC0-1.0.
+

claude2md-0.1.dist-info/RECORD

@@ -0,0 +1,7 @@
+claude2md/__init__.py,sha256=OJSOeXiscuXAlUj-FghIIMEy9rLuEl5kCxM7XmBtEow,10421
+claude2md/__main__.py,sha256=UoVzR7sYrF46j-rTtHoq_NglBMXpHD_jYSFqyzK1Qlw,66
+claude2md-0.1.dist-info/entry_points.txt,sha256=MzOYtfNQ6ecjaOUBi9ptkTEadyqMn9pYmzGAC6F8buA,44
+claude2md-0.1.dist-info/licenses/LICENSE,sha256=ogEPNDSH0_dhiv_lT3ifVIdgIzHAqNA_SemnxUfPBJk,7048
+claude2md-0.1.dist-info/WHEEL,sha256=G2gURzTEtmeR8nrdXUJfNiB3VYVxigPQ-bEQujpNiNs,82
+claude2md-0.1.dist-info/METADATA,sha256=PpSVMGUSV5muRtfTOCsRr2stXAgHRreuj1K3dkp7VbU,5710
+claude2md-0.1.dist-info/RECORD,,

claude2md-0.1.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: flit 3.12.0
+Root-Is-Purelib: true
+Tag: py3-none-any

claude2md-0.1.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+claude2md=claude2md:main
+

claude2md-0.1.dist-info/licenses/LICENSE

@@ -0,0 +1,121 @@
+Creative Commons Legal Code
+
+CC0 1.0 Universal
+
+    CREATIVE COMMONS CORPORATION IS NOT A LAW FIRM AND DOES NOT PROVIDE
+    LEGAL SERVICES. DISTRIBUTION OF THIS DOCUMENT DOES NOT CREATE AN
+    ATTORNEY-CLIENT RELATIONSHIP. CREATIVE COMMONS PROVIDES THIS
+    INFORMATION ON AN "AS-IS" BASIS. CREATIVE COMMONS MAKES NO WARRANTIES
+    REGARDING THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS
+    PROVIDED HEREUNDER, AND DISCLAIMS LIABILITY FOR DAMAGES RESULTING FROM
+    THE USE OF THIS DOCUMENT OR THE INFORMATION OR WORKS PROVIDED
+    HEREUNDER.
+
+Statement of Purpose
+
+The laws of most jurisdictions throughout the world automatically confer
+exclusive Copyright and Related Rights (defined below) upon the creator
+and subsequent owner(s) (each and all, an "owner") of an original work of
+authorship and/or a database (each, a "Work").
+
+Certain owners wish to permanently relinquish those rights to a Work for
+the purpose of contributing to a commons of creative, cultural and
+scientific works ("Commons") that the public can reliably and without fear
+of later claims of infringement build upon, modify, incorporate in other
+works, reuse and redistribute as freely as possible in any form whatsoever
+and for any purposes, including without limitation commercial purposes.
+These owners may contribute to the Commons to promote the ideal of a free
+culture and the further production of creative, cultural and scientific
+works, or to gain reputation or greater distribution for their Work in
+part through the use and efforts of others.
+
+For these and/or other purposes and motivations, and without any
+expectation of additional consideration or compensation, the person
+associating CC0 with a Work (the "Affirmer"), to the extent that he or she
+is an owner of Copyright and Related Rights in the Work, voluntarily
+elects to apply CC0 to the Work and publicly distribute the Work under its
+terms, with knowledge of his or her Copyright and Related Rights in the
+Work and the meaning and intended legal effect of CC0 on those rights.
+
+1. Copyright and Related Rights. A Work made available under CC0 may be
+protected by copyright and related or neighboring rights ("Copyright and
+Related Rights"). Copyright and Related Rights include, but are not
+limited to, the following:
+
+  i. the right to reproduce, adapt, distribute, perform, display,
+     communicate, and translate a Work;
+ ii. moral rights retained by the original author(s) and/or performer(s);
+iii. publicity and privacy rights pertaining to a person's image or
+     likeness depicted in a Work;
+ iv. rights protecting against unfair competition in regards to a Work,
+     subject to the limitations in paragraph 4(a), below;
+  v. rights protecting the extraction, dissemination, use and reuse of data
+     in a Work;
+ vi. database rights (such as those arising under Directive 96/9/EC of the
+     European Parliament and of the Council of 11 March 1996 on the legal
+     protection of databases, and under any national implementation
+     thereof, including any amended or successor version of such
+     directive); and
+vii. other similar, equivalent or corresponding rights throughout the
+     world based on applicable law or treaty, and any national
+     implementations thereof.
+
+2. Waiver. To the greatest extent permitted by, but not in contravention
+of, applicable law, Affirmer hereby overtly, fully, permanently,
+irrevocably and unconditionally waives, abandons, and surrenders all of
+Affirmer's Copyright and Related Rights and associated claims and causes
+of action, whether now known or unknown (including existing as well as
+future claims and causes of action), in the Work (i) in all territories
+worldwide, (ii) for the maximum duration provided by applicable law or
+treaty (including future time extensions), (iii) in any current or future
+medium and for any number of copies, and (iv) for any purpose whatsoever,
+including without limitation commercial, advertising or promotional
+purposes (the "Waiver"). Affirmer makes the Waiver for the benefit of each
+member of the public at large and to the detriment of Affirmer's heirs and
+successors, fully intending that such Waiver shall not be subject to
+revocation, rescission, cancellation, termination, or any other legal or
+equitable action to disrupt the quiet enjoyment of the Work by the public
+as contemplated by Affirmer's express Statement of Purpose.
+
+3. Public License Fallback. Should any part of the Waiver for any reason
+be judged legally invalid or ineffective under applicable law, then the
+Waiver shall be preserved to the maximum extent permitted taking into
+account Affirmer's express Statement of Purpose. In addition, to the
+extent the Waiver is so judged Affirmer hereby grants to each affected
+person a royalty-free, non transferable, non sublicensable, non exclusive,
+irrevocable and unconditional license to exercise Affirmer's Copyright and
+Related Rights in the Work (i) in all territories worldwide, (ii) for the
+maximum duration provided by applicable law or treaty (including future
+time extensions), (iii) in any current or future medium and for any number
+of copies, and (iv) for any purpose whatsoever, including without
+limitation commercial, advertising or promotional purposes (the
+"License"). The License shall be deemed effective as of the date CC0 was
+applied by Affirmer to the Work. Should any part of the License for any
+reason be judged legally invalid or ineffective under applicable law, such
+partial invalidity or ineffectiveness shall not invalidate the remainder
+of the License, and in such case Affirmer hereby affirms that he or she
+will not (i) exercise any of his or her remaining Copyright and Related
+Rights in the Work or (ii) assert any associated claims and causes of
+action with respect to the Work, in either case contrary to Affirmer's
+express Statement of Purpose.
+
+4. Limitations and Disclaimers.
+
+ a. No trademark or patent rights held by Affirmer are waived, abandoned,
+    surrendered, licensed or otherwise affected by this document.
+ b. Affirmer offers the Work as-is and makes no representations or
+    warranties of any kind concerning the Work, express, implied,
+    statutory or otherwise, including without limitation warranties of
+    title, merchantability, fitness for a particular purpose, non
+    infringement, or the absence of latent or other defects, accuracy, or
+    the present or absence of errors, whether or not discoverable, all to
+    the greatest extent permissible under applicable law.
+ c. Affirmer disclaims responsibility for clearing rights of other persons
+    that may apply to the Work or any use thereof, including without
+    limitation any person's Copyright and Related Rights in the Work.
+    Further, Affirmer disclaims responsibility for obtaining any necessary
+    consents, permissions or other rights required for any use of the
+    Work.
+ d. Affirmer understands and acknowledges that Creative Commons is not a
+    party to this document and has no duty or obligation with respect to
+    this CC0 or use of the Work.