ocrd 3.6.0__tar.gz → 3.7.0__tar.gz

{ocrd-3.6.0/src/ocrd.egg-info → ocrd-3.7.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ocrd
-Version: 3.6.0
+Version: 3.7.0
 Summary: OCR-D framework
 Author-email: Konstantin Baierer <unixprog@gmail.com>
 License: Apache License 2.0
@@ -37,7 +37,7 @@ Requires-Dist: opencv-python-headless
 Requires-Dist: paramiko
 Requires-Dist: pika>=1.2.0
 Requires-Dist: Pillow>=7.2.0
-Requires-Dist: pydantic==1.*
+Requires-Dist: pydantic>=2.0.0
 Requires-Dist: python-magic
 Requires-Dist: python-multipart
 Requires-Dist: pyyaml
@@ -220,12 +220,6 @@ Also contains the command line tool `ocrd`.
 
 See [README for `ocrd`](./README_ocrd.md) for further information.
 
-## bash library
-
-Builds a bash script that can be sourced by other bash scripts to create OCRD-compliant CLI.
-
-See [README for `bashlib`](./README_bashlib.md) for further information.
-
 ## Testing
 
 Download assets (`make assets`)

{ocrd-3.6.0 → ocrd-3.7.0}/README.md

@@ -170,12 +170,6 @@ Also contains the command line tool `ocrd`.
 
 See [README for `ocrd`](./README_ocrd.md) for further information.
 
-## bash library
-
-Builds a bash script that can be sourced by other bash scripts to create OCRD-compliant CLI.
-
-See [README for `bashlib`](./README_bashlib.md) for further information.
-
 ## Testing
 
 Download assets (`make assets`)

ocrd-3.7.0/VERSION

@@ -0,0 +1 @@
+3.7.0

{ocrd-3.6.0 → ocrd-3.7.0}/requirements.txt

@@ -26,7 +26,7 @@ opencv-python-headless
 paramiko
 pika>=1.2.0
 Pillow >= 7.2.0
-pydantic==1.*
+pydantic >= 2.0.0
 python-magic
 python-multipart
 pyyaml

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd/cli/__init__.py

@@ -32,12 +32,11 @@ from ..decorators import ocrd_loglevel
 from .ocrd_tool import ocrd_tool_cli
 from .workspace import workspace_cli
 from .process import process_cli
-from .bashlib import bashlib_cli
 from .validate import validate_cli
 from .resmgr import resmgr_cli
 from .zip import zip_cli
-from .log import log_cli
 from .network import network_cli
+from .bashlib import bashlib_cli
 
 
 __all__ = ['cli']
@@ -117,9 +116,8 @@ def cli(**kwargs):  # pylint: disable=unused-argument
 cli.add_command(ocrd_tool_cli)
 cli.add_command(workspace_cli)
 cli.add_command(process_cli)
-cli.add_command(bashlib_cli)
 cli.add_command(zip_cli)
 cli.add_command(validate_cli)
-cli.add_command(log_cli)
 cli.add_command(resmgr_cli)
 cli.add_command(network_cli)
+cli.add_command(bashlib_cli)

ocrd-3.7.0/src/ocrd/cli/bashlib.py

@@ -0,0 +1,42 @@
+"""
+OCR-D CLI: bash library
+
+.. click:: ocrd.cli.bashlib:bashlib_cli
+    :prog: ocrd bashlib
+    :nested: full
+
+"""
+
+# WARNING: bashlib processors have been deprecated as of v3 of the OCR-D/core API
+#          and will be removed in v3.7.0. We retain the `ocrd bashlib` CLI only
+#          to not break the `ocrd bashlib filename` command, which is used in CD
+#          scripts to get the `share` directory of the core installation. 
+
+import click
+from ocrd.constants import BASHLIB_FILENAME
+
+# ----------------------------------------------------------------------
+# ocrd bashlib
+# ----------------------------------------------------------------------
+
+
+@click.group('bashlib')
+def bashlib_cli():
+    """
+    Work with bash library
+    """
+
+# ----------------------------------------------------------------------
+# ocrd bashlib filename
+# ----------------------------------------------------------------------
+
+
+@bashlib_cli.command('filename')
+def bashlib_filename():
+    """
+    Dump the bash library filename for sourcing by shell scripts
+
+    For functions exported by bashlib, see `<../../README.md>`_
+    """
+    print(BASHLIB_FILENAME)
+

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd/mets_server.py

@@ -258,12 +258,12 @@ class ClientSideOcrdMets:
 
     def add_agent(self, **kwargs):
         if not self.multiplexing_mode:
-            return self.session.request("POST", f"{self.url}/agent", json=OcrdAgentModel.create(**kwargs).dict())
+            return self.session.request("POST", f"{self.url}/agent", json=OcrdAgentModel.create(**kwargs).model_dump())
         else:
             self.session.request(
                 "POST",
                 self.url,
-                json=MpxReq.add_agent(self.ws_dir_path, OcrdAgentModel.create(**kwargs).dict())
+                json=MpxReq.add_agent(self.ws_dir_path, OcrdAgentModel.create(**kwargs).model_dump())
             ).json()
             return OcrdAgentModel.create(**kwargs)
 
@@ -305,7 +305,7 @@ class ClientSideOcrdMets:
             mimetype=mimetype, url=url, local_filename=local_filename
         )
         # add force+ignore
-        kwargs = {**kwargs, **data.dict()}
+        kwargs = {**kwargs, **data.model_dump()}
 
         if not self.multiplexing_mode:
             r = self.session.request("POST", f"{self.url}/file", data=kwargs)
@@ -530,7 +530,7 @@ class OcrdMetsServer:
 
         @app.post(path='/agent', response_model=OcrdAgentModel)
         async def add_agent(agent: OcrdAgentModel):
-            kwargs = agent.dict()
+            kwargs = agent.model_dump()
             kwargs['_type'] = kwargs.pop('type')
             workspace.mets.add_agent(**kwargs)
             response = agent
@@ -575,7 +575,7 @@ class OcrdMetsServer:
                 local_filename=local_filename
             )
             # Add to workspace
-            kwargs = file_resource.dict()
+            kwargs = file_resource.model_dump()
             workspace.add_file(**kwargs, force=force)
             response = file_resource
             self.log.debug(f"POST /file -> {response.__dict__}")

{ocrd-3.6.0 → ocrd-3.7.0/src/ocrd.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ocrd
-Version: 3.6.0
+Version: 3.7.0
 Summary: OCR-D framework
 Author-email: Konstantin Baierer <unixprog@gmail.com>
 License: Apache License 2.0
@@ -37,7 +37,7 @@ Requires-Dist: opencv-python-headless
 Requires-Dist: paramiko
 Requires-Dist: pika>=1.2.0
 Requires-Dist: Pillow>=7.2.0
-Requires-Dist: pydantic==1.*
+Requires-Dist: pydantic>=2.0.0
 Requires-Dist: python-magic
 Requires-Dist: python-multipart
 Requires-Dist: pyyaml
@@ -220,12 +220,6 @@ Also contains the command line tool `ocrd`.
 
 See [README for `ocrd`](./README_ocrd.md) for further information.
 
-## bash library
-
-Builds a bash script that can be sourced by other bash scripts to create OCRD-compliant CLI.
-
-See [README for `bashlib`](./README_bashlib.md) for further information.
-
 ## Testing
 
 Download assets (`make assets`)

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd.egg-info/SOURCES.txt

@@ -1,7 +1,6 @@
 LICENSE
 MANIFEST.in
 README.md
-README_bashlib.md
 README_ocrd.md
 README_ocrd_modelfactory.md
 README_ocrd_models.md
@@ -13,7 +12,6 @@ pyproject.toml
 requirements.txt
 src/ocrd/__init__.py
 src/ocrd/constants.py
-src/ocrd/lib.bash
 src/ocrd/mets_server.py
 src/ocrd/ocrd-all-tool.json
 src/ocrd/resolver.py
@@ -31,7 +29,6 @@ src/ocrd.egg-info/requires.txt
 src/ocrd.egg-info/top_level.txt
 src/ocrd/cli/__init__.py
 src/ocrd/cli/bashlib.py
-src/ocrd/cli/log.py
 src/ocrd/cli/network.py
 src/ocrd/cli/ocrd_tool.py
 src/ocrd/cli/process.py

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd.egg-info/requires.txt

@@ -22,7 +22,7 @@ opencv-python-headless
 paramiko
 pika>=1.2.0
 Pillow>=7.2.0
-pydantic==1.*
+pydantic>=2.0.0
 python-magic
 python-multipart
 pyyaml

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd_network/models/job.py

@@ -13,7 +13,7 @@ class PYJobInput(BaseModel):
     workspace_id: Optional[str] = None
     description: Optional[str] = None
     input_file_grps: List[str]
-    output_file_grps: Optional[List[str]]
+    output_file_grps: Optional[List[str]] = None
     page_id: Optional[str] = None
     parameters: dict = {}  # Always set to empty dict when None, otherwise it fails ocr-d-validation
     result_queue_name: Optional[str] = None
@@ -23,8 +23,8 @@ class PYJobInput(BaseModel):
     # If set, specifies a list of job ids this job depends on
     depends_on: Optional[List[str]] = None
 
-    class Config:
-        schema_extra = {
+    model_config = {
+        'json_schema_extra': {
             'example': {
                 'path_to_mets': '/path/to/mets.xml',
                 'description': 'The description of this execution',
@@ -34,6 +34,7 @@ class PYJobInput(BaseModel):
                 'parameters': {}
             }
         }
+    }
 
 
 class PYJobOutput(BaseModel):
@@ -42,12 +43,12 @@ class PYJobOutput(BaseModel):
     job_id: str
     processor_name: str
     state: JobState = JobState.unset
-    path_to_mets: Optional[str]
-    workspace_id: Optional[str]
+    path_to_mets: Optional[str] = None
+    workspace_id: Optional[str] = None
     input_file_grps: List[str]
-    output_file_grps: Optional[List[str]]
+    output_file_grps: Optional[List[str]] = None
     page_id: Optional[str] = None
-    log_file_path: Optional[str]
+    log_file_path: Optional[str] = None
 
 
 class DBProcessorJob(Document):
@@ -55,22 +56,22 @@ class DBProcessorJob(Document):
     """
     job_id: str
     processor_name: str
-    path_to_mets: Optional[str]
-    workspace_id: Optional[str]
-    description: Optional[str]
+    path_to_mets: Optional[str] = None
+    workspace_id: Optional[str] = None
+    description: Optional[str] = None
     state: JobState = JobState.unset
     input_file_grps: List[str]
-    output_file_grps: Optional[List[str]]
-    page_id: Optional[str]
-    parameters: Optional[dict]
-    depends_on: Optional[List[str]]
-    result_queue_name: Optional[str]
-    callback_url: Optional[str]
-    internal_callback_url: Optional[str]
-    start_time: Optional[datetime]
-    end_time: Optional[datetime]
-    exec_time: Optional[str]
-    log_file_path: Optional[str]
+    output_file_grps: Optional[List[str]] = None
+    page_id: Optional[str] = None
+    parameters: Optional[dict] = None
+    depends_on: Optional[List[str]] = None
+    result_queue_name: Optional[str] = None
+    callback_url: Optional[str] = None
+    internal_callback_url: Optional[str] = None
+    start_time: Optional[datetime] = None
+    end_time: Optional[datetime] = None
+    exec_time: Optional[str] = None
+    log_file_path: Optional[str] = None
 
     class Settings:
         use_enum_values = True
@@ -99,9 +100,9 @@ class PYWorkflowJobOutput(BaseModel):
     page_id: str
     page_wise: bool = False
     job_id: str
-    path_to_mets: Optional[str]
-    workspace_id: Optional[str]
-    description: Optional[str]
+    path_to_mets: Optional[str] = None
+    workspace_id: Optional[str] = None
+    description: Optional[str] = None
 
 
 class DBWorkflowJob(Document):
@@ -114,10 +115,10 @@ class DBWorkflowJob(Document):
     # key: page_id
     # value: List of and processing job ids sorted in dependency order
     processing_job_ids: Dict
-    path_to_mets: Optional[str]
-    workspace_id: Optional[str]
-    description: Optional[str]
-    workflow_callback_url: Optional[str]
+    path_to_mets: Optional[str] = None
+    workspace_id: Optional[str] = None
+    description: Optional[str] = None
+    workflow_callback_url: Optional[str] = None
 
     class Settings:
         use_enum_values = True

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd_network/models/messages.py

@@ -11,8 +11,8 @@ class PYResultMessage(BaseModel):
     path_to_mets: Optional[str] = None
     workspace_id: Optional[str] = None
 
-    class Config:
-        schema_extra = {
+    model_config = {
+        "json_schema_extra": {
             "example": {
                 "job_id": "d8e36726-ed28-5476-b83c-bc31d2eecf1f",
                 "state": JobState.success,
@@ -20,3 +20,4 @@ class PYResultMessage(BaseModel):
                 "workspace_id": "c7f25615-fc17-4365-a74d-ad20e1ddbd0e"
             }
         }
+    }

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd_network/models/workspace.py

@@ -25,10 +25,10 @@ class DBWorkspace(Document):
     workspace_mets_path: str
     ocrd_identifier: str
     bagit_profile_identifier: str
-    ocrd_base_version_checksum: Optional[str]
-    ocrd_mets: Optional[str]
-    bag_info_adds: Optional[dict]
-    mets_server_url: Optional[str]
+    ocrd_base_version_checksum: Optional[str] = None
+    ocrd_mets: Optional[str] = None
+    bag_info_adds: Optional[dict] = None
+    mets_server_url: Optional[str] = None
     deleted: bool = False
 
     class Settings:

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd_utils/logging.py

@@ -75,6 +75,9 @@ _ocrdLevel2pythonLevel = {
 
 
 def tf_disable_interactive_logs():
+    """
+    Disable the interactive logging of tf/keras and set the log level to error or higher
+    """
     try:
         from os import environ  # pylint: disable=import-outside-toplevel
         # This env variable must be set before importing from Keras

{ocrd-3.6.0 → ocrd-3.7.0}/src/ocrd_utils/os.py

@@ -171,7 +171,7 @@ def list_all_resources(executable, moduled=None, xdg_data_home=None):
                    # files over directories; but we want data files only
                    # todo: more code and cache exclusion patterns!
                    ['*.py', '*.py[cod]', '*~', 'ocrd-tool.json',
-                    'environment.pickle', 'resource_list.yml', 'lib.bash']):
+                    'environment.pickle', 'resource_list.yml']):
                 continue
             candidates.append(resource)
     # recurse once

ocrd-3.6.0/README_bashlib.md

@@ -1,177 +0,0 @@
-# bashlib
-
-> Builds a bash script that can be sourced by other bash scripts to create OCRD-compliant CLI.
-
-For example:
-
-    source `ocrd bashlib filename`
-    declare -A NAMESPACES MIMETYPES
-    eval NAMESPACES=( `ocrd bashlib constants NAMESPACES` )
-    echo ${NAMESPACES[page]}
-    eval MIMETYPE_PAGE=( `ocrd bashlib constants MIMETYPE_PAGE` )
-    echo $MIMETYPE_PAGE
-    eval MIMETYPES=( `ocrd bashlib constants EXT_TO_MIME` )
-    echo ${MIMETYPES[.jpg]}
-
-
-<!-- BEGIN-MARKDOWN-TOC -->
-* [bashlib CLI](#bashlib-cli)
-* [bashlib API](#bashlib-api)
-* [`ocrd__raise`](#ocrd__raise)
-* [`ocrd__log`](#ocrd__log)
-* [`ocrd__minversion`](#ocrd__minversion)
-* [`ocrd__dumpjson`](#ocrd__dumpjson)
-* [`ocrd__resolve_resource`](#ocrd__resolve_resource)
-* [`ocrd__show_resource`](#ocrd__show_resource)
-* [`ocrd__list_resources`](#ocrd__list_resources)
-* [`ocrd__usage`](#ocrd__usage)
-* [`ocrd__parse_argv`](#ocrd__parse_argv)
-<!-- END-MARKDOWN-TOC -->
-
-## bashlib CLI
-
-See [CLI usage](https://ocr-d.de/core/api/ocrd/ocrd.cli.bashlib.html)
-
-## bashlib API
-
-### `ocrd__raise`
-
-Raise an error and exit.
-
-### `ocrd__log`
-
-Delegate logging to [`ocrd log`](#ocrd-cli)
-
-### `ocrd__minversion`
-
-Ensure minimum version
-
-### `ocrd__dumpjson`
-
-Output ocrd-tool.json content verbatim.
-
-Requires `$OCRD_TOOL_JSON` and `$OCRD_TOOL_NAME` to be set:
-
-```sh
-export OCRD_TOOL_JSON=/path/to/ocrd-tool.json
-export OCRD_TOOL_NAME=ocrd-foo-bar
-```
-
-(Which you automatically get from [`ocrd__wrap`](#ocrd__wrap).)
-
-### `ocrd__resolve_resource`
-
-Output given resource file's path.
-
-### `ocrd__show_resource`
-
-Output given resource file's content.
-
-### `ocrd__list_resources`
-
-Output all resource files' names.
-
-### `ocrd__usage`
-
-Print help on CLI usage.
-
-### `ocrd__parse_argv`
-
-Parses arguments according to [OCR-D CLI](https://ocr-d.de/en/spec/cli).
-In doing so, depending on the values passed to it, may delegate to …
-- [`ocrd__raise`](#ocrd__raise) and exit (if something went wrong)
-- [`ocrd__usage`](#ocrd__usage) and exit
-- [`ocrd__dumpjson`](#ocrd__dumpjson) and exit
-- [`ocrd__show_resource`](#ocrd__show_resource) and exit
-- [`ocrd__list_resources`](#ocrd__list_resources) and exit
-- [`ocrd validate tasks`](#ocrd-cli) and return
-
-Expects an associative array ("hash"/"dict") **`ocrd__argv`** to be predefined:
-
-    declare -A ocrd__argv=()
-
-This will be filled by the parser along the following keys:
-- `overwrite`: whether `--overwrite` is enabled
-- `profile`: whether `--profile` is enabled
-- `profile_file`: the argument of `--profile-file`
-- `log_level`: the argument of `--log-level`
-- `mets_server_url`: the argument of `--mets-server-url` argument
-- `mets_file`: absolute path of the `--mets` argument
-- `working_dir`: absolute path of the `--working-dir` argument or the parent of `mets_file`
-- `page_id`: the argument of `--page-id`
-- `input_file_grp`: the argument of `--input-file-grp`
-- `output_file_grp`: the argument of `--output-file-grp`
-
-Moreover, there will be an associative array **`params`**
-with the fully validated and default-expanded runtime values of the `ocrd-tool.json` parameters.
-
-### `ocrd__wrap`
-
-Parses an [ocrd-tool.json](https://ocr-d.de/en/spec/ocrd_tool) for a specific `tool` (i.e. processor `executable`).
-
-Delegates to …
-- [`ocrd__parse_argv`](#ocrd__parse_argv), creating the `ocrd__argv` associative array
-- [`ocrd bashlib input-files`](#ocrd-cli), creating the data structures used by [`ocrd__input_file`](#ocrd__input_file)
-
-Usage: `ocrd__wrap PATH/TO/OCRD-TOOL.JSON EXECUTABLE ARGS`
-
-For example:
-
-    ocrd__wrap $SHAREDIR/ocrd-tool.json ocrd-olena-binarize "$@"
-    ...
-
-### `ocrd__input_file`
-
-(Requires [`ocrd__wrap`](#ocrd__wrap) to have been run first.)
-
-Access information on the input files according to the parsed CLI arguments:
-- their file `url` (or local file path)
-- their file `ID`
-- their `mimetype`
-- their `pageId`
-- their proposed corresponding `outputFileId` (generated from `${ocrd__argv[output__file_grp]}` and input file `ID`)
-
-Usage: `ocrd__input_file NR KEY`
-
-For example:
-
-    pageId=`ocrd__input_file 3 pageId`
-
-To be used in a **loop over all selected pages**:
-
-    for ((n=0; n<${#ocrd__files[*]}; n++)); do
-        local in_fpath=($(ocrd__input_file $n url))
-        local in_id=($(ocrd__input_file $n ID))
-        local in_mimetype=($(ocrd__input_file $n mimetype))
-        local in_pageId=($(ocrd__input_file $n pageId))
-        local out_id=$(ocrd__input_file $n outputFileId)
-        local out_fpath="${ocrd__argv[output_file_grp]}/${out_id}.xml
-
-        # process $in_fpath to $out_fpath ...
-
-        declare -a options
-        if [ -n "$in_pageId" ]; then
-            options=( -g $in_pageId )
-        else
-            options=()
-        fi
-        if [[ "${ocrd__argv[overwrite]}" == true ]];then
-            options+=( --force )
-        fi
-        options+=( -G ${ocrd__argv[output_file_grp]}
-                   -m $MIMETYPE_PAGE -i "$out_id"
-                   "$out_fpath" )
-        ocrd -l ${ocrd__argv[log_level]} workspace -d ${ocrd__argv[working_dir]} add "${options[@]}"
-
-> **Note**: If the `--input-file-grp` is **multi-valued** (N fileGrps separated by commas),
-> then usage is similar:
-> * The function `ocrd__input_file` can be used, but
->   its results will be **lists** (delimited by whitespace and surrounded by single quotes),
->   e.g. `[url]='file1.xml file2.xml' [ID]='id_file1 id_file2' [mimetype]='application/vnd.prima.page+xml image/tiff' ...`.
-> * Therefore its results should be encapsulated in a (non-associative) **array variable**
->   and without extra quotes, e.g. `in_file=($(ocrd__input_file 3 url))`, or as shown above.
-> * This will yield the first fileGrp's results on index 0,
->   which in bash will always be the same as if you referenced the array without index
->   (so code does not need to be changed much), e.g. `test -f $in_file` which equals `test -f ${in_file[0]}`.
-> * Additional fileGrps will have to be fetched from higher indexes, e.g. `test -f ${in_file[1]}`.
-

ocrd-3.6.0/VERSION

@@ -1 +0,0 @@
-3.6.0