Skip to content

Module: repo_manager

Utility functions for uploading/downloading repositories to/from Arweave. Each repository contains multiple artifact files.

Repository files (artifacts) are logically grouped together via a Manifest file, which maps individual transaction data to named files.

When uploading a repo directory, a version mapping dictionary file is expected to be provided. The mapping should contain a map of filename to version tag. The version tag is useful if a specific version of a file is meant to be downloaded. If no mapping is specified, version will be an empty string.

NotFinalizedException

Bases: Exception

Exception raised when a transaction is not yet finalized.

Source code in src/ritual_arweave/repo_manager.py
class NotFinalizedException(Exception):
    """Exception raised when a transaction is not yet finalized."""

    pass

RepoManager

Bases: FileManager

Source code in src/ritual_arweave/repo_manager.py
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
class RepoManager(FileManager):
    def download_artifact_file(
        self,
        repo_id: Union[ArRepoId, str],
        file_name: str,
        version: Optional[str] = None,
        force_download: bool = False,
        base_path: str | Path = ".",
    ) -> Path:
        """Downloads a specific artifact from Arweave.

        Args:
            repo_id (Union[ArRepoId, str]): id of the repo, if provided as a string, the
                format must be of the form `owner`/`name`. Where `owner` is the wallet
                address of the uploader and `name` is the repository's name.
            file_name (str): name of artifact
            version (Optional[str], optional): Version of file. Defaults to None.
                If none specified, will fetch the latest version.
            force_download (bool, optional): If True, will download file even if it
                already exists at that path. Defaults to False.
            base_path (str, optional): path to download file to. Defaults to "".

        Raises:
            ValueError: if wallet file path is not specified or wallet file is not found.

        Returns:
            Path: path to downloaded file
        """
        if isinstance(repo_id, str):
            repo_id = ArRepoId.from_str(repo_id)

        base = Path(base_path)
        os.makedirs(base, exist_ok=True)
        owners = [repo_id.owner]

        file_version_str = (
            ""
            if not version
            else """
            {
                name: "File-Version",
                values: ["%s"]
            },
        """
            % version
        )
        query_str = """
        query {
            transactions(
                sort:HEIGHT_DESC,
                owners: %s,
                tags: [
                    {
                        name: "App-Name",
                        values: ["Ritual"]
                    },
                    %s
                    {
                        name: "File-Name",
                        values: ["%s"]
                    },
                    {
                        name: "Repo-Name",
                        values: ["%s"]
                    }
                ])
            {
                edges {
                    node {
                        block {
                            id
                            timestamp
                        }
                        id
                        owner {
                            address
                        }
                        tags {
                            name
                            value
                        }
                    }
                }
            }
        }
        """ % (
            json.dumps(owners),
            file_version_str,
            file_name,
            repo_id.name,
        )
        log.debug(query_str)

        file_path: Path = base.joinpath(file_name)

        res = self.peer.graphql(query_str)
        transactions = res["data"]["transactions"]["edges"]
        transactions.sort(reverse=True, key=edge_unix_ts)

        if len(transactions) == 0:
            raise ValueError(
                f"Could not find any matching artifacts for: "
                f"({repo_id}, {file_name}, {version or 'latest'})"
            )

        transaction = transactions[0]

        tx_metadata: dict[str, Any] = transaction["node"]

        tx_id = tx_metadata["id"]

        if force_download or not self.file_exists(str(file_path), tx_id):
            log.info(f"downloading {tx_metadata}")
            return self.download(str(file_path), tx_id)
        else:
            log.info(f"not downloading {tx_metadata} because it already exists")
            return Path(file_path)

    def upload_repo(
        self,
        name: str,
        path: Path | str,
        version_mapping_file: Optional[str] = None,
        version_mapping: Optional[Dict[str, str]] = None,
        extra_file_tags: Optional[Dict[str, Tags]] = None,
    ) -> UploadArweaveRepoResult:
        """
        Uploads a repo directory to Arweave. For every repository upload, a manifest
        mapping is created.

        Args:
            name (str): Name of the repository. Once uploaded, the repo will be
                accessible via the repo Id: `owner/name`. Where `owner` is the wallet
                address of the uploader and `name` is the repository's name.
            path (Path | str): Path to the directory containing the repository files.
            version_mapping_file (str): Path to a json dict file mapping file names to
                specific versions. If a specific mapping is found, the File-Version
                attribute is tagged with the value. This is to facilitate uploading and
                downloading version specific files.
            version_mapping (dict[str, str]): Dictionary mapping file names to specific
                versions. If a specific mapping is found, the File-Version attribute is
                tagged with the value. This is to facilitate uploading and downloading
                version specific files. If provided, this will override the version_file.
            extra_file_tags (dict[str, Tags]): Dictionary mapping file names to
                additional tags to be added to the file. This is useful for adding
                additional metadata to each file.

        Raises:
            ValueError: if wallet file path is not specified or wallet file is not found.

        Returns:
            UploadArweaveRepoResult: Result of the upload containing repo_id,
            transaction_id, and manifest_url.
        """

        # path to load files from
        _path: Path = Path(path)

        # load all sub-paths in this path
        p = _path.glob("**/*")

        # get timestamp to tag files with
        timestamp = time.time()

        # filter out simlinks and non-files
        files = [x for x in p if x.is_file()]

        _version_mapping = {}
        if version_mapping:
            _version_mapping = version_mapping
        elif version_mapping_file:
            with open(version_mapping_file, "r") as vf:
                _version_mapping = json.load(vf)
        self.logger(f"using mapping {_version_mapping}")

        # keep track of entries via a manifest
        manifest_dict: dict[str, str] = {}

        ritual_tags: Tags = {
            "App-Name": "Ritual",
            "App-Version": VERSION,
            "Unix-Time": str(timestamp),
            "Repo-Name": str(name),
        }

        for f in files:
            rel_path = os.path.relpath(f, _path)

            self.logger(f"looking at {f} ({rel_path}) Size: {os.path.getsize(f)}")

            content_type = (
                guess
                if (guess := mimetypes.guess_type(f)[0])
                else "application/octet-stream"
            )
            file_extra_tags = (
                extra_file_tags.get(rel_path, {}) if extra_file_tags else {}
            )

            tags_dict: Tags = {
                **file_extra_tags,
                "Content-Type": content_type,
                "File-Version": _version_mapping.get(str(rel_path), "0.0.0"),
                "File-Name": rel_path,
                "File-SHA256": get_sha256_digest(str(f)),
                **ritual_tags,
            }

            self.logger(f"uploading: {f} with tags: {tags_dict}")

            tx = self.upload(f, tags_dict)

            # we are done uploading the whole file, keep track if filename -> tx.id
            manifest_dict[str(os.path.relpath(f, _path))] = tx.id
            self.logger(f"uploaded file {f} with id {tx.id} and tags {tags_dict}")

        # we create a manifest of all the files to their transactions
        m = Manifest(manifest_dict)

        # upload the manifest
        t = Transaction(self.wallet, peer=self.peer, data=m.tobytes())

        t.add_tags(
            {
                "Content-Type": "application/x.arweave-manifest+json",
                "Type": "manifest",
                **ritual_tags,
            }
        )

        t.sign()
        t.send()

        self.logger(f"uploaded manifest with tx id {t.id}")

        return UploadArweaveRepoResult(
            repo_id=ArRepoId(owner=self.wallet.address, name=name),
            transaction_id=t.id,
            manifest_url=f"{t.api_url}/{t.id}",
        )

    def download_file_in_repo(
        self,
        repo_id: Union[ArRepoId, str],
        file_name: str,
        base_path: str | Path = Path("."),
    ) -> Path:
        """
        Downloads a file from a repo on Arweave.

        Args:
            repo_id: Arweave repo id, in the format of `owner`/`name` where `owner` is
                the wallet address of the uploader and `name` is the repository's name.
            file_name: name of the file to download.
            base_path: path to download the file to.

        Returns:
            path to the downloaded file.

        """

        base_path = Path(base_path)
        manifest = self.get_repo_manifest(repo_id)
        file_tid = manifest["paths"][file_name]["id"]
        return self.download(base_path / file_name, file_tid)

    def get_repo_manifest(self, repo_id: Union[ArRepoId, str]) -> dict[str, Any]:
        """
        Get the manifest of a repo from Arweave.
        Args:
            repo_id: Arweave repo id, in the format of `owner`/`name` where `owner` is
                the wallet address of the uploader and `name` is the repository's name.

        Returns:
            dict: manifest of the repo.

        """

        if isinstance(repo_id, str):
            repo_id = ArRepoId.from_str(repo_id)
        owners = [repo_id.owner]
        query_str = """
        query {
            transactions(
                sort:HEIGHT_DESC,
                owners: %s,
                tags: [
                    {
                        name: "App-Name",
                        values: ["Ritual"]
                    },
                    {
                        name: "Repo-Name",
                        values: ["%s"]
                    },
                    {
                        name: "Type",
                        values: ["manifest"]
                    }
                ]
            )
            {
                edges {
                    node {
                        block {
                            id
                            timestamp
                        }
                        id
                        owner {
                            address
                        }
                        tags {
                            name
                            value
                        }
                    }
                }
            }
        }
        """ % (
            json.dumps(owners),
            repo_id.name,
        )

        # self.logger(query_str)
        log.info("getting first query")
        try:
            res = self.peer.graphql(query_str)
        except HTTPError as e:
            raise ValueError(f"Error querying repo manifests: {e}")
        log.info("done getting first query")

        # get latest Manifest

        manifests = res["data"]["transactions"]["edges"]
        manifests.sort(reverse=True, key=edge_unix_ts)
        self.logger(f"found {len(manifests)} manifests for {repo_id}")

        if len(manifests) == 0:
            raise ValueError("Could not find any matching repo manifests from query.")

        manifest = manifests[0]

        tx_id = manifest["node"]["id"]

        # download manifest data
        self.logger(f"found manifest {manifest}")

        log.info("getting manifest")
        try:
            m = json.loads(self.peer.tx_data(tx_id))
        except Exception as e:
            log.info("Exception while getting manifest")
            r = requests.get(f"{self.peer.api_url}/tx/{tx_id}")
            if r.status_code == 202:
                raise NotFinalizedException(
                    f"Manifest {tx_id} is still being mined. Please try again later."
                )
            raise ValueError(f"Error fetching manifest data: {e}")
        log.info("done getting manifest")

        self.logger(f"loaded manifest {m}")
        return cast(dict[str, Any], m)

    def download_repo(
        self,
        repo_id: Union[ArRepoId, str],
        base_path: str | Path = Path("."),
        force_download: bool = False,
    ) -> List[Path]:
        """Downloads a repo from Arweave to a given directory.

        Args:
            repo_id (Union[ArRepoId, str]): id of the repo, if provided as a string, the
                format must be of the form `owner`/`name`. Where `owner` is the wallet
                address of the uploader and `name` is the respository's name.
            base_path (str | Path, optional): path to download files to. Defaults to ".".
            force_download (bool, optional): If True, will download files even if they
                already exist. Defaults to False.

        Raises:
            ValueError: if wallet file path is not specified or wallet file is not found.
            ValueError: if matching repo manifest not found
            NotFinalizedException: if the manifest is still being mined

        Returns:
            list[str]: downloaded file paths
        """

        manifest = self.get_repo_manifest(repo_id)

        base = Path(base_path)
        base.mkdir(parents=True, exist_ok=True)

        paths = []

        with ThreadPoolExecutor() as executor:
            for pathname, tid in manifest["paths"].items():
                file_tid: str = tid["id"]
                joined_path: Path = base.joinpath(pathname)

                # check if file exists
                if force_download or not self.file_exists(str(joined_path), file_tid):
                    st = time.time()
                    self.logger(f"downloading file {pathname} for {file_tid}")
                    paths.append(
                        executor.submit(self.download, str(joined_path), file_tid)
                    )
                    self.logger(f"downloaded in {time.time() - st} sec: {joined_path}")
                else:
                    self.logger(
                        f"Path {joined_path} already exists and will not be downloaded. "
                        + "Please remove it or use --force_download flag."
                    )

        return [p.result() for p in paths]

download_artifact_file(repo_id, file_name, version=None, force_download=False, base_path='.')

Downloads a specific artifact from Arweave.

Parameters:

Name Type Description Default
repo_id Union[ArRepoId, str]

id of the repo, if provided as a string, the format must be of the form owner/name. Where owner is the wallet address of the uploader and name is the repository's name.

required
file_name str

name of artifact

required
version Optional[str]

Version of file. Defaults to None. If none specified, will fetch the latest version.

None
force_download bool

If True, will download file even if it already exists at that path. Defaults to False.

False
base_path str

path to download file to. Defaults to "".

'.'

Raises:

Type Description
ValueError

if wallet file path is not specified or wallet file is not found.

Returns:

Name Type Description
Path Path

path to downloaded file

Source code in src/ritual_arweave/repo_manager.py
def download_artifact_file(
    self,
    repo_id: Union[ArRepoId, str],
    file_name: str,
    version: Optional[str] = None,
    force_download: bool = False,
    base_path: str | Path = ".",
) -> Path:
    """Downloads a specific artifact from Arweave.

    Args:
        repo_id (Union[ArRepoId, str]): id of the repo, if provided as a string, the
            format must be of the form `owner`/`name`. Where `owner` is the wallet
            address of the uploader and `name` is the repository's name.
        file_name (str): name of artifact
        version (Optional[str], optional): Version of file. Defaults to None.
            If none specified, will fetch the latest version.
        force_download (bool, optional): If True, will download file even if it
            already exists at that path. Defaults to False.
        base_path (str, optional): path to download file to. Defaults to "".

    Raises:
        ValueError: if wallet file path is not specified or wallet file is not found.

    Returns:
        Path: path to downloaded file
    """
    if isinstance(repo_id, str):
        repo_id = ArRepoId.from_str(repo_id)

    base = Path(base_path)
    os.makedirs(base, exist_ok=True)
    owners = [repo_id.owner]

    file_version_str = (
        ""
        if not version
        else """
        {
            name: "File-Version",
            values: ["%s"]
        },
    """
        % version
    )
    query_str = """
    query {
        transactions(
            sort:HEIGHT_DESC,
            owners: %s,
            tags: [
                {
                    name: "App-Name",
                    values: ["Ritual"]
                },
                %s
                {
                    name: "File-Name",
                    values: ["%s"]
                },
                {
                    name: "Repo-Name",
                    values: ["%s"]
                }
            ])
        {
            edges {
                node {
                    block {
                        id
                        timestamp
                    }
                    id
                    owner {
                        address
                    }
                    tags {
                        name
                        value
                    }
                }
            }
        }
    }
    """ % (
        json.dumps(owners),
        file_version_str,
        file_name,
        repo_id.name,
    )
    log.debug(query_str)

    file_path: Path = base.joinpath(file_name)

    res = self.peer.graphql(query_str)
    transactions = res["data"]["transactions"]["edges"]
    transactions.sort(reverse=True, key=edge_unix_ts)

    if len(transactions) == 0:
        raise ValueError(
            f"Could not find any matching artifacts for: "
            f"({repo_id}, {file_name}, {version or 'latest'})"
        )

    transaction = transactions[0]

    tx_metadata: dict[str, Any] = transaction["node"]

    tx_id = tx_metadata["id"]

    if force_download or not self.file_exists(str(file_path), tx_id):
        log.info(f"downloading {tx_metadata}")
        return self.download(str(file_path), tx_id)
    else:
        log.info(f"not downloading {tx_metadata} because it already exists")
        return Path(file_path)

download_file_in_repo(repo_id, file_name, base_path=Path('.'))

Downloads a file from a repo on Arweave.

Parameters:

Name Type Description Default
repo_id Union[ArRepoId, str]

Arweave repo id, in the format of owner/name where owner is the wallet address of the uploader and name is the repository's name.

required
file_name str

name of the file to download.

required
base_path str | Path

path to download the file to.

Path('.')

Returns:

Type Description
Path

path to the downloaded file.

Source code in src/ritual_arweave/repo_manager.py
def download_file_in_repo(
    self,
    repo_id: Union[ArRepoId, str],
    file_name: str,
    base_path: str | Path = Path("."),
) -> Path:
    """
    Downloads a file from a repo on Arweave.

    Args:
        repo_id: Arweave repo id, in the format of `owner`/`name` where `owner` is
            the wallet address of the uploader and `name` is the repository's name.
        file_name: name of the file to download.
        base_path: path to download the file to.

    Returns:
        path to the downloaded file.

    """

    base_path = Path(base_path)
    manifest = self.get_repo_manifest(repo_id)
    file_tid = manifest["paths"][file_name]["id"]
    return self.download(base_path / file_name, file_tid)

download_repo(repo_id, base_path=Path('.'), force_download=False)

Downloads a repo from Arweave to a given directory.

Parameters:

Name Type Description Default
repo_id Union[ArRepoId, str]

id of the repo, if provided as a string, the format must be of the form owner/name. Where owner is the wallet address of the uploader and name is the respository's name.

required
base_path str | Path

path to download files to. Defaults to ".".

Path('.')
force_download bool

If True, will download files even if they already exist. Defaults to False.

False

Raises:

Type Description
ValueError

if wallet file path is not specified or wallet file is not found.

ValueError

if matching repo manifest not found

NotFinalizedException

if the manifest is still being mined

Returns:

Type Description
List[Path]

list[str]: downloaded file paths

Source code in src/ritual_arweave/repo_manager.py
def download_repo(
    self,
    repo_id: Union[ArRepoId, str],
    base_path: str | Path = Path("."),
    force_download: bool = False,
) -> List[Path]:
    """Downloads a repo from Arweave to a given directory.

    Args:
        repo_id (Union[ArRepoId, str]): id of the repo, if provided as a string, the
            format must be of the form `owner`/`name`. Where `owner` is the wallet
            address of the uploader and `name` is the respository's name.
        base_path (str | Path, optional): path to download files to. Defaults to ".".
        force_download (bool, optional): If True, will download files even if they
            already exist. Defaults to False.

    Raises:
        ValueError: if wallet file path is not specified or wallet file is not found.
        ValueError: if matching repo manifest not found
        NotFinalizedException: if the manifest is still being mined

    Returns:
        list[str]: downloaded file paths
    """

    manifest = self.get_repo_manifest(repo_id)

    base = Path(base_path)
    base.mkdir(parents=True, exist_ok=True)

    paths = []

    with ThreadPoolExecutor() as executor:
        for pathname, tid in manifest["paths"].items():
            file_tid: str = tid["id"]
            joined_path: Path = base.joinpath(pathname)

            # check if file exists
            if force_download or not self.file_exists(str(joined_path), file_tid):
                st = time.time()
                self.logger(f"downloading file {pathname} for {file_tid}")
                paths.append(
                    executor.submit(self.download, str(joined_path), file_tid)
                )
                self.logger(f"downloaded in {time.time() - st} sec: {joined_path}")
            else:
                self.logger(
                    f"Path {joined_path} already exists and will not be downloaded. "
                    + "Please remove it or use --force_download flag."
                )

    return [p.result() for p in paths]

get_repo_manifest(repo_id)

Get the manifest of a repo from Arweave. Args: repo_id: Arweave repo id, in the format of owner/name where owner is the wallet address of the uploader and name is the repository's name.

Returns:

Name Type Description
dict dict[str, Any]

manifest of the repo.

Source code in src/ritual_arweave/repo_manager.py
def get_repo_manifest(self, repo_id: Union[ArRepoId, str]) -> dict[str, Any]:
    """
    Get the manifest of a repo from Arweave.
    Args:
        repo_id: Arweave repo id, in the format of `owner`/`name` where `owner` is
            the wallet address of the uploader and `name` is the repository's name.

    Returns:
        dict: manifest of the repo.

    """

    if isinstance(repo_id, str):
        repo_id = ArRepoId.from_str(repo_id)
    owners = [repo_id.owner]
    query_str = """
    query {
        transactions(
            sort:HEIGHT_DESC,
            owners: %s,
            tags: [
                {
                    name: "App-Name",
                    values: ["Ritual"]
                },
                {
                    name: "Repo-Name",
                    values: ["%s"]
                },
                {
                    name: "Type",
                    values: ["manifest"]
                }
            ]
        )
        {
            edges {
                node {
                    block {
                        id
                        timestamp
                    }
                    id
                    owner {
                        address
                    }
                    tags {
                        name
                        value
                    }
                }
            }
        }
    }
    """ % (
        json.dumps(owners),
        repo_id.name,
    )

    # self.logger(query_str)
    log.info("getting first query")
    try:
        res = self.peer.graphql(query_str)
    except HTTPError as e:
        raise ValueError(f"Error querying repo manifests: {e}")
    log.info("done getting first query")

    # get latest Manifest

    manifests = res["data"]["transactions"]["edges"]
    manifests.sort(reverse=True, key=edge_unix_ts)
    self.logger(f"found {len(manifests)} manifests for {repo_id}")

    if len(manifests) == 0:
        raise ValueError("Could not find any matching repo manifests from query.")

    manifest = manifests[0]

    tx_id = manifest["node"]["id"]

    # download manifest data
    self.logger(f"found manifest {manifest}")

    log.info("getting manifest")
    try:
        m = json.loads(self.peer.tx_data(tx_id))
    except Exception as e:
        log.info("Exception while getting manifest")
        r = requests.get(f"{self.peer.api_url}/tx/{tx_id}")
        if r.status_code == 202:
            raise NotFinalizedException(
                f"Manifest {tx_id} is still being mined. Please try again later."
            )
        raise ValueError(f"Error fetching manifest data: {e}")
    log.info("done getting manifest")

    self.logger(f"loaded manifest {m}")
    return cast(dict[str, Any], m)

upload_repo(name, path, version_mapping_file=None, version_mapping=None, extra_file_tags=None)

Uploads a repo directory to Arweave. For every repository upload, a manifest mapping is created.

Parameters:

Name Type Description Default
name str

Name of the repository. Once uploaded, the repo will be accessible via the repo Id: owner/name. Where owner is the wallet address of the uploader and name is the repository's name.

required
path Path | str

Path to the directory containing the repository files.

required
version_mapping_file str

Path to a json dict file mapping file names to specific versions. If a specific mapping is found, the File-Version attribute is tagged with the value. This is to facilitate uploading and downloading version specific files.

None
version_mapping dict[str, str]

Dictionary mapping file names to specific versions. If a specific mapping is found, the File-Version attribute is tagged with the value. This is to facilitate uploading and downloading version specific files. If provided, this will override the version_file.

None
extra_file_tags dict[str, Tags]

Dictionary mapping file names to additional tags to be added to the file. This is useful for adding additional metadata to each file.

None

Raises:

Type Description
ValueError

if wallet file path is not specified or wallet file is not found.

Returns:

Name Type Description
UploadArweaveRepoResult UploadArweaveRepoResult

Result of the upload containing repo_id,

UploadArweaveRepoResult

transaction_id, and manifest_url.

Source code in src/ritual_arweave/repo_manager.py
def upload_repo(
    self,
    name: str,
    path: Path | str,
    version_mapping_file: Optional[str] = None,
    version_mapping: Optional[Dict[str, str]] = None,
    extra_file_tags: Optional[Dict[str, Tags]] = None,
) -> UploadArweaveRepoResult:
    """
    Uploads a repo directory to Arweave. For every repository upload, a manifest
    mapping is created.

    Args:
        name (str): Name of the repository. Once uploaded, the repo will be
            accessible via the repo Id: `owner/name`. Where `owner` is the wallet
            address of the uploader and `name` is the repository's name.
        path (Path | str): Path to the directory containing the repository files.
        version_mapping_file (str): Path to a json dict file mapping file names to
            specific versions. If a specific mapping is found, the File-Version
            attribute is tagged with the value. This is to facilitate uploading and
            downloading version specific files.
        version_mapping (dict[str, str]): Dictionary mapping file names to specific
            versions. If a specific mapping is found, the File-Version attribute is
            tagged with the value. This is to facilitate uploading and downloading
            version specific files. If provided, this will override the version_file.
        extra_file_tags (dict[str, Tags]): Dictionary mapping file names to
            additional tags to be added to the file. This is useful for adding
            additional metadata to each file.

    Raises:
        ValueError: if wallet file path is not specified or wallet file is not found.

    Returns:
        UploadArweaveRepoResult: Result of the upload containing repo_id,
        transaction_id, and manifest_url.
    """

    # path to load files from
    _path: Path = Path(path)

    # load all sub-paths in this path
    p = _path.glob("**/*")

    # get timestamp to tag files with
    timestamp = time.time()

    # filter out simlinks and non-files
    files = [x for x in p if x.is_file()]

    _version_mapping = {}
    if version_mapping:
        _version_mapping = version_mapping
    elif version_mapping_file:
        with open(version_mapping_file, "r") as vf:
            _version_mapping = json.load(vf)
    self.logger(f"using mapping {_version_mapping}")

    # keep track of entries via a manifest
    manifest_dict: dict[str, str] = {}

    ritual_tags: Tags = {
        "App-Name": "Ritual",
        "App-Version": VERSION,
        "Unix-Time": str(timestamp),
        "Repo-Name": str(name),
    }

    for f in files:
        rel_path = os.path.relpath(f, _path)

        self.logger(f"looking at {f} ({rel_path}) Size: {os.path.getsize(f)}")

        content_type = (
            guess
            if (guess := mimetypes.guess_type(f)[0])
            else "application/octet-stream"
        )
        file_extra_tags = (
            extra_file_tags.get(rel_path, {}) if extra_file_tags else {}
        )

        tags_dict: Tags = {
            **file_extra_tags,
            "Content-Type": content_type,
            "File-Version": _version_mapping.get(str(rel_path), "0.0.0"),
            "File-Name": rel_path,
            "File-SHA256": get_sha256_digest(str(f)),
            **ritual_tags,
        }

        self.logger(f"uploading: {f} with tags: {tags_dict}")

        tx = self.upload(f, tags_dict)

        # we are done uploading the whole file, keep track if filename -> tx.id
        manifest_dict[str(os.path.relpath(f, _path))] = tx.id
        self.logger(f"uploaded file {f} with id {tx.id} and tags {tags_dict}")

    # we create a manifest of all the files to their transactions
    m = Manifest(manifest_dict)

    # upload the manifest
    t = Transaction(self.wallet, peer=self.peer, data=m.tobytes())

    t.add_tags(
        {
            "Content-Type": "application/x.arweave-manifest+json",
            "Type": "manifest",
            **ritual_tags,
        }
    )

    t.sign()
    t.send()

    self.logger(f"uploaded manifest with tx id {t.id}")

    return UploadArweaveRepoResult(
        repo_id=ArRepoId(owner=self.wallet.address, name=name),
        transaction_id=t.id,
        manifest_url=f"{t.api_url}/{t.id}",
    )

UploadArweaveRepoResult

Bases: BaseModel

Model to represent the result of a repository upload.

Source code in src/ritual_arweave/repo_manager.py
class UploadArweaveRepoResult(BaseModel):
    """Model to represent the result of a repository upload."""

    repo_id: ArRepoId
    transaction_id: str
    manifest_url: str