ocrd 3.10.1__tar.gz → 3.12.1__tar.gz

{ocrd-3.10.1/src/ocrd.egg-info → ocrd-3.12.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocrd
-Version: 3.10.1
+Version: 3.12.1
 Summary: OCR-D framework
 Author-email: Konstantin Baierer <unixprog@gmail.com>
 License: Apache License 2.0

ocrd-3.12.1/VERSION

@@ -0,0 +1 @@
+3.12.1

{ocrd-3.10.1 → ocrd-3.12.1}/src/ocrd/cli/workspace.py

@@ -6,7 +6,7 @@ OCR-D CLI: workspace management
     :nested: full
 """
 from os import rmdir, unlink
-from os.path import dirname, relpath, normpath, exists, join, isabs, isdir
+from os.path import abspath, dirname, relpath, normpath, exists, join, isabs, isdir
 from pathlib import Path
 from json import loads, dumps
 import sys
@@ -47,7 +47,7 @@ class WorkspaceCtx():
     def workspace(self):
         return Workspace(
             self.resolver,
-            directory=self.directory,
+            self.directory,
             mets_basename=self.mets_basename,
             automatic_backup=self.automatic_backup,
             mets_server_url=self.mets_server_url,
@@ -85,7 +85,7 @@ def workspace_cli(ctx, directory, mets, mets_basename, mets_server_url, backup):
     initLogging()
     ctx.obj = WorkspaceCtx(
         directory,
-        mets_url=mets,
+        mets,
         mets_basename=mets_basename,
         mets_server_url=mets_server_url,
         automatic_backup=backup
@@ -254,22 +254,26 @@ def workspace_add_file(ctx, file_grp, file_id, mimetype, page_id, ignore, check_
     log.debug("Adding '%s'", fname)
     local_filename = None
     if not (fname.startswith('http://') or fname.startswith('https://')):
-        if not fname.startswith(ctx.directory):
-            if not isabs(fname) and exists(join(ctx.directory, fname)):
+        assert isabs(ctx.directory)
+        # try to resolve relative to workspace or CWD
+        if not isabs(fname):
+            if exists(join(ctx.directory, fname)):
                 fname = join(ctx.directory, fname)
             else:
-                log.debug("File '%s' is not in workspace, copying", fname)
-                try:
-                    fname = ctx.resolver.download_to_directory(ctx.directory, fname, subdir=file_grp)
-                except FileNotFoundError:
-                    if check_file_exists:
-                        log.error("File '%s' does not exist, halt execution!" % fname)
-                        sys.exit(1)
-        if check_file_exists and not exists(fname):
-            log.error("File '%s' does not exist, halt execution!" % fname)
-            sys.exit(1)
+                fname = abspath(fname)
         if fname.startswith(ctx.directory):
+            if check_file_exists and not exists(fname):
+                log.error("File '%s' does not exist, halt execution!" % fname)
+                sys.exit(1)
             fname = relpath(fname, ctx.directory)
+        else:
+            log.debug("File '%s' is not in workspace, copying", fname)
+            try:
+                fname = ctx.resolver.download_to_directory(ctx.directory, fname, subdir=file_grp)
+            except FileNotFoundError:
+                if check_file_exists:
+                    log.error("File '%s' does not exist, halt execution!" % fname)
+                    sys.exit(1)
         local_filename = fname
 
     if not page_id:
@@ -299,8 +303,8 @@ def workspace_add_file(ctx, file_grp, file_id, mimetype, page_id, ignore, check_
 @click.option('-g', '--page-id', help="physical page ID of the file", required=False)
 @click.option('-i', '--file-id', help="ID of the file. If not provided, derive from fileGrp and filename", required=False)
 @click.option('-u', '--url', help="Remote URL of the file", required=False)
-@click.option('-l', '--local-filename', help="Local filesystem path in the workspace directory "
-              "(copied from source file if different)", required=False)
+@click.option('-l', '--local-filename', help="Relative filesystem path in the workspace directory "
+              "(copied from source path if different)", required=False)
 @click.option('-G', '--file-grp', help="File group USE of the file", required=True)
 @click.option('-n', '--dry-run', help="Don't actually do anything to the METS or filesystem, just preview",
               default=False, is_flag=True)
@@ -414,38 +418,45 @@ def workspace_cli_bulk_add(ctx, regex, mimetype, page_id, file_id, url, local_fi
                 raise ValueError(f"OcrdFile attribute '{param_name}' unset ({file_dict})")
             for group_name in group_dict:
                 file_dict[param_name] = file_dict[param_name].replace('{{ %s }}' % group_name, group_dict[group_name])
+            if '{{ ' in file_dict[param_name]:
+                log.warning("Probably failed to match a capture group in %s: '%s'", param_name, file_dict[param_name])
 
         # Where to copy from
         if src_path_option:
-            src_path = src_path_option
             for group_name in group_dict:
-                src_path = src_path.replace('{{ %s }}' % group_name, group_dict[group_name])
-            srcpath = Path(src_path)
+                src_path_option = src_path_option.replace('{{ %s }}' % group_name, group_dict[group_name])
+            if '{{ ' in src_path_option:
+                log.warning("Probably failed to match a capture group in source path: '%s'", src_path_option)
+            src_path = Path(src_path_option)
         else:
-            srcpath = file_path
+            src_path = file_path
 
         # derive --file-id from filename if not --file-id not explicitly set
         if not file_id:
-            id_field = srcpath.stem if file_path != srcpath else file_path.stem
+            id_field = src_path.stem if file_path != src_path else file_path.stem
             file_dict['file_id'] = safe_filename('%s_%s' % (file_dict['file_grp'], id_field))
         if not mimetype:
             try:
-                file_dict['mimetype'] = EXT_TO_MIME[srcpath.suffix]
+                file_dict['mimetype'] = EXT_TO_MIME[src_path.suffix]
             except KeyError:
                 log.error("Cannot guess MIME type from extension '%s' for '%s'. "
-                          "Set --mimetype explicitly" % (srcpath.suffix, srcpath))
+                          "Set --mimetype explicitly" % (src_path.suffix, src_path))
 
         # copy files if src != url
+        src_path = src_path.resolve().absolute()
         if local_filename_is_src:
-            file_dict['local_filename'] = srcpath
-        else:
-            destpath = Path(workspace.directory, file_dict['local_filename'])
-            if srcpath != destpath and not destpath.exists():
-                log.info("cp '%s' '%s'", srcpath, destpath)
-                if not dry_run:
-                    if not destpath.parent.is_dir():
-                        destpath.parent.mkdir()
-                    destpath.write_bytes(srcpath.read_bytes())
+            if str(src_path).startswith(workspace.directory):
+                file_dict['local_filename'] = src_path.relative_to(workspace.directory)
+            else:
+                file_dict['local_filename'] = src_path.name # will be copied-in below
+
+        dest_path = Path(workspace.directory, file_dict['local_filename'])
+        if src_path != dest_path and not dest_path.exists():
+            log.info("cp '%s' '%s'", src_path, dest_path)
+            if not dry_run:
+                if not dest_path.parent.is_dir():
+                    dest_path.parent.mkdir()
+                dest_path.write_bytes(src_path.read_bytes())
 
         # Add to workspace (or not)
         fileGrp = file_dict.pop('file_grp')
@@ -465,8 +476,7 @@ def workspace_cli_bulk_add(ctx, regex, mimetype, page_id, file_id, url, local_fi
 @workspace_cli.command('find')
 @mets_find_options
 @click.option('-k', '--output-field', help="Output field. Repeat for multiple fields, will be joined with tab",
-              default=['local_filename'],
-              show_default=True,
+              default=['local_filename'], show_default=True,
               multiple=True,
               type=click.Choice([
                   'url',
@@ -869,7 +879,7 @@ def merge(ctx, overwrite, force, copy_files, filegrp_mapping, fileid_mapping, pa
     the same semantics as in ``ocrd workspace find``, see ``ocrd workspace find --help``
     for an explanation.
     """
-    mets_path = Path(mets_path)
+    mets_path = Path(mets_path).absolute()
     if filegrp_mapping:
         filegrp_mapping = loads(filegrp_mapping)
     assert not ctx.mets_server_url, \

{ocrd-3.10.1 → ocrd-3.12.1}/src/ocrd/mets_server.py

@@ -418,7 +418,8 @@ class OcrdMetsServer:
     def create_process(mets_server_url: str, ws_dir_path: str, log_file: str) -> int:
         sub_process = Popen(
             args=["ocrd", "workspace", "-U", f"{mets_server_url}", "-d", f"{ws_dir_path}", "server", "start"],
-            stdout=open(file=log_file, mode="w"), stderr=open(file=log_file, mode="a"), cwd=ws_dir_path,
+            stdout=open(file=log_file, mode="w"), stderr=open(file=log_file, mode="a"),
+            #cwd=ws_dir_path, # rs: if relative, this will cause wrong path resolution in the subprocess
             shell=False, universal_newlines=True, start_new_session=True
         )
         # Wait for the mets server to start
@@ -446,7 +447,7 @@ class OcrdMetsServer:
                 Path(self.url).unlink()
 
     def startup(self):
-        self.log.info("Configuring the METS Server")
+        self.log.info("Configuring the METS Server for %s", self.workspace)
 
         workspace = self.workspace
 

{ocrd-3.10.1 → ocrd-3.12.1}/src/ocrd/processor/base.py

@@ -539,7 +539,7 @@ class Processor():
                     self._base_logger.info("limiting page timeout from %d to %d sec", max_seconds, self.max_page_seconds)
                     max_seconds = self.max_page_seconds
 
-                if isinstance(workspace.mets, ClientSideOcrdMets):
+                if max_workers > 1:
                     executor_cls = ProcessPoolExecutor
                     log_queue = mp.get_context('fork').Queue()
                 else:
@@ -553,7 +553,7 @@ class Processor():
                     initializer=_page_worker_set_ctxt,
                     initargs=(self, log_queue),
                 )
-                if isinstance(workspace.mets, ClientSideOcrdMets):
+                if max_workers > 1:
                     assert executor.active # ensure pre-forking
                     # forward messages from log queue (in subprocesses) to all root handlers
                     log_listener = logging.handlers.QueueListener(log_queue, *logging.root.handlers,

{ocrd-3.10.1 → ocrd-3.12.1/src/ocrd.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocrd
-Version: 3.10.1
+Version: 3.12.1
 Summary: OCR-D framework
 Author-email: Konstantin Baierer <unixprog@gmail.com>
 License: Apache License 2.0

{ocrd-3.10.1 → ocrd-3.12.1}/src/ocrd_network/constants.py

@@ -2,8 +2,6 @@ from enum import Enum
 
 DOCKER_IMAGE_MONGO_DB = "mongo"
 DOCKER_IMAGE_RABBIT_MQ = "rabbitmq:4.2-management"
-# These feature flags are required by default to use the newer version
-DOCKER_RABBIT_MQ_FEATURES = "quorum_queue,implicit_default_bindings,classic_mirrored_queue_version"
 
 NETWORK_PROTOCOLS = ["http://", "https://"]
 OCRD_ALL_TOOL_JSON = "ocrd-all-tool.json"

{ocrd-3.10.1 → ocrd-3.12.1}/src/ocrd_network/processing_server.py

@@ -388,7 +388,7 @@ class ProcessingServer(FastAPI):
         self.cache_processing_requests.update_request_counter(workspace_key=workspace_key, by_value=0)
 
         # This check is done to return early in case a workspace_id is provided
-        # but the abs mets path cannot be queried from the DB
+        # but the mets path cannot be queried from the DB
         request_mets_path = await validate_and_return_mets_path(self.log, data)
 
         page_ids = expand_page_ids(data.page_id)

{ocrd-3.10.1 → ocrd-3.12.1}/src/ocrd_network/runtime_data/network_services.py

@@ -2,7 +2,7 @@ from __future__ import annotations
 from logging import Logger
 from typing import Any, Dict, List, Optional, Union
 
-from ..constants import DOCKER_IMAGE_MONGO_DB, DOCKER_IMAGE_RABBIT_MQ, DOCKER_RABBIT_MQ_FEATURES
+from ..constants import DOCKER_IMAGE_MONGO_DB, DOCKER_IMAGE_RABBIT_MQ
 from ..database import verify_mongodb_available
 from ..rabbitmq_utils import verify_rabbitmq_available
 from .connection_clients import create_docker_client
@@ -136,9 +136,7 @@ class DataRabbitMQ(DataNetworkService):
             env = [
                 # The default credentials to be used by the processing workers
                 f"RABBITMQ_DEFAULT_USER={rmq_user}",
-                f"RABBITMQ_DEFAULT_PASS={rmq_password}",
-                # These feature flags are required by default to use the newer version
-                f"RABBITMQ_FEATURE_FLAGS={DOCKER_RABBIT_MQ_FEATURES}"
+                f"RABBITMQ_DEFAULT_PASS={rmq_password}"
             ]
         if not ports_mapping:
             # 5672, 5671 - used by AMQP 0-9-1 and AMQP 1.0 clients without and with TLS

ocrd-3.10.1/VERSION

@@ -1 +0,0 @@
-3.10.1