Merge branch 'dev' into f-rocrate-converter

CHANGELOG.md

@@ -20,6 +20,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   ```
 - Support for Python 3.13
 - ROCrateConverter, ELNFileConverter and ROCrateEntityConverter for crawling ROCrate and .eln files
+- `max_log_level` parameter to `logging.configure_server_side_logging`
+  to control the server-side debuglog's verboosity, and an optional
+  `sss_max_log_level` parameter to `crawler_main` to control the SSS
+  loglevel separately from the global `debug` option.
 
 ### Changed ###
 
@@ -30,6 +34,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - if `value` starts with '+', collection mode is "list".
   - if `value` starts with '*', collection mode is "multiproperty".
   - in all other cases, collection mode is "single".
+- The default server-side scrippting debug level is now controlled by
+  the global `debug` option by default and set to log level `INFO` in
+  case of `debug=False`. The previous behavior can be restored by
+  calling `crawler_main` with `sss_max_log_level=logging.DEBUG`.
 
 ### Deprecated ###
 
@@ -39,6 +47,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Added better error message for some cases of broken converter and
   record definitions.
+- [#108](https://gitlab.com/linkahead/linkahead-crawler/-/issues/108)
+  Too verbose server-side scripting logs that could lead to high disk
+  usage.
 
 ### Security ###
 

src/caoscrawler/crawl.py

@@ -1020,6 +1020,7 @@ def crawler_main(crawled_directory_path: str,
                  restricted_path: Optional[list[str]] = None,
                  remove_prefix: Optional[str] = None,
                  add_prefix: Optional[str] = None,
+                 sss_max_log_level: Optional[int] = None,
                  ):
     """
 
@@ -1053,6 +1054,12 @@ def crawler_main(crawled_directory_path: str,
     add_prefix : Optional[str]
         Add the given prefix to file paths.
         See docstring of '_fix_file_paths' for more details.
+    sss_max_log_level : Optional[int]
+        If given, set the maximum log level of the server-side
+        scripting log separately from the general ``debug`` option. If
+        None is given, the maximum sss log level will be determined
+        from the value of ``debug``: ``logging.INFO`` if ``debug`` is
+        False, ``logging.DEBUG`` if ``debug`` is True.
 
     Returns
     -------
@@ -1063,7 +1070,11 @@ def crawler_main(crawled_directory_path: str,
         crawler = Crawler(securityMode=securityMode)
 
         if "SHARED_DIR" in os.environ:  # setup logging and reporting if serverside execution
-            userlog_public, htmluserlog_public, debuglog_public = configure_server_side_logging()
+            if sss_max_log_level is None:
+                sss_max_log_level = logging.DEBUG if debug else logging.INFO
+            userlog_public, htmluserlog_public, debuglog_public = configure_server_side_logging(
+                max_log_level=sss_max_log_level
+            )
             # TODO make this optional
             _create_status_record(
                 get_shared_resource_link(get_config_setting("public_host_url"), htmluserlog_public),

src/caoscrawler/logging.py

@@ -26,23 +26,40 @@ from caosadvancedtools.serverside.helper import get_shared_filename
 import sys
 
 
-def configure_server_side_logging():
+def configure_server_side_logging(max_log_level: int = logging.INFO):
     """
     Set logging up to save one plain debugging log file, one plain info log
     file (for users) and a stdout stream with messages wrapped in html elements
 
     returns the path to the file with debugging output
+
+    Parameters
+    ----------
+    max_log_level : int, optional
+        The maximum log level to use for SSS-logs. Default is
+        ``logging.INFO``.
+
+    Returns
+    -------
+    userlog_public, htmluserlog_public, debuglog_public: str
+        Public paths of the respective log files.
     """
     adv_logger = logging.getLogger("caosadvancedtools")
-    adv_logger.setLevel(level=logging.DEBUG)
+    # The max_<level> variables will be used to set the logger levels
+    # to the respective maximum of intended level and max_log_level,
+    # effectively cutting off logging above the specified
+    # max_log_level.
+    max_info = max(logging.INFO, max_log_level)
+    max_debug = max(logging.DEBUG, max_log_level)
+    adv_logger.setLevel(level=max_debug)
 
     cr_logger = logging.getLogger("caoscrawler")
-    cr_logger.setLevel(level=logging.DEBUG)
+    cr_logger.setLevel(level=max_debug)
 
     userlog_public, userlog_internal = get_shared_filename("userlog.txt")
 
     root_logger = logging.getLogger()
-    root_logger.setLevel(level=logging.INFO)
+    root_logger.setLevel(level=max_info)
 
     # this is a log file with INFO level for the user
     user_file_handler = logging.FileHandler(filename=userlog_internal)