Add --server to override any baseurl

Author: lfoppiano

Readme.md

@@ -55,7 +55,7 @@ usage: grobid_client [-h] [--input INPUT] [--output OUTPUT] [--config CONFIG]
                      [--n N] [--generateIDs] [--consolidate_header]
                      [--consolidate_citations] [--include_raw_citations]
                      [--include_raw_affiliations] [--force] [--teiCoordinates]
-                     [--verbose] [--flavor FLAVOR]
+                     [--verbose] [--flavor FLAVOR] [--server SERVER]
                      service
 
 Client for GROBID services
@@ -75,6 +75,7 @@ optional arguments:
                         (optional)
   --config CONFIG       path to the config file, default is ./config.json
   --n N                 concurrency for service usage
+  --server SERVER       GROBID server URL (default: http://localhost:8070)
   --generateIDs         generate random xml:id to textual XML elements of the
                         result files
   --consolidate_header  call GROBID with consolidation of the metadata
@@ -132,6 +133,15 @@ The following command example will process all the PDF files present in the inpu
 > grobid_client --input ~/tmp/in2 --output ~/tmp/out --teiCoordinates --segmentSentences processFulltextDocument
 ```
 
+To use a different GROBID server (e.g., a hosted service), use the `--server` argument:
+
+```console
+> grobid_client --server https://lfoppiano-grobid.hf.space --input ~/tmp/in2 --output ~/tmp/out processFulltextDocument
+```
+
+> [!NOTE]
+> The `--server` argument will override the server URL specified in the config file. If both are provided, the CLI argument takes precedence.
+
 The file `example.py` gives an example of usage as a library, from a another python script. 
 
 ## Using the client in your python
@@ -141,16 +151,24 @@ Import and call the client as follow:
 ```python
 from grobid_client.grobid_client import GrobidClient
 
+# Using default localhost server
 client = GrobidClient(config_path="./config.json")
 client.process("processFulltextDocument", "/mnt/data/covid/pdfs", n=20)
+
+# Using a custom server
+client = GrobidClient(grobid_server="https://lfoppiano-grobid.hf.space", config_path="./config.json")
+client.process("processFulltextDocument", "/mnt/data/covid/pdfs", n=20)
 ```
 
 See also `example.py`.
 
 ## Configuration of the client
 
 > [!TIP]
-> from version 0.0.12 the `config.json` will be optional, by default the client will connect to the local server (`http://localhost:8070`). 
+> from version 0.0.12 the `config.json` will be optional, by default the client will connect to the local server (`http://localhost:8070`).
+
+> [!NOTE]
+> When using the CLI, the `--server` argument will override the `grobid_server` value in the config file. This allows you to use a config file for most settings while easily switching servers via command line. 
 
 There are a few parameters that can be set with the `config.json` file. 
 

example.py

@@ -1,5 +1,10 @@
 from grobid_client.grobid_client import GrobidClient
 
 if __name__ == "__main__":
+    # Example with default localhost server
     client = GrobidClient(config_path="./config.json")
     client.process("processFulltextDocument", "./resources/test_pdf", output="./resources/test_out/", consolidate_citations=True, tei_coordinates=True, force=True)
+    
+    # Example with custom server (uncomment to use)
+    # client = GrobidClient(grobid_server="https://lfoppiano-grobid.hf.space", config_path="./config.json")
+    # client.process("processFulltextDocument", "./resources/test_pdf", output="./resources/test_out/", consolidate_citations=True, tei_coordinates=True, force=True)

grobid_client/grobid_client.py

@@ -81,9 +81,33 @@ def __init__(
             }
         }
 
+        # Store constructor parameters to ensure they take precedence over config file
+        # We'll use these to override config file values after loading
+        self._constructor_params = {
+            'grobid_server': grobid_server,
+            'batch_size': batch_size,
+            'coordinates': coordinates,
+            'sleep_time': sleep_time,
+            'timeout': timeout
+        }
+
         # Load config first (which may override logging settings)
         if config_path:
             self._load_config(config_path)
+            
+        # Ensure constructor parameters take precedence over config file values
+        # This ensures CLI arguments override config file values
+        # Only override if the parameter was explicitly provided (not the default)
+        if grobid_server != 'http://localhost:8070':
+            self.config['grobid_server'] = grobid_server
+        if batch_size != 1000:
+            self.config['batch_size'] = batch_size
+        if coordinates is not None:  # coordinates has a complex default, so check if explicitly provided
+            self.config['coordinates'] = coordinates
+        if sleep_time != 5:
+            self.config['sleep_time'] = sleep_time
+        if timeout != 180:
+            self.config['timeout'] = timeout
 
         # Configure logging based on config
         self._configure_logging()
@@ -718,6 +742,11 @@ def main():
         default=None,
         help="Define the flavor to be used for the fulltext extraction",
     )
+    parser.add_argument(
+        "--server",
+        default="http://localhost:8070",
+        help="GROBID server URL (default: http://localhost:8070)",
+    )
 
     args = parser.parse_args()
 
@@ -736,7 +765,7 @@ def main():
 
     # Initialize GrobidClient which will configure logging based on config.json
     try:
-        client = GrobidClient(config_path=config_path)
+        client = GrobidClient(grobid_server=args.server, config_path=config_path)
         # Now use the client's logger for all subsequent logging
         logger = client.logger
     except ServerUnavailableException as e:

tests/test_grobid_client.py

@@ -43,7 +43,7 @@ def test_init_default_values(self, mock_configure_logging, mock_test_server):
         assert client.config['grobid_server'] == 'http://localhost:8070'
         assert client.config['batch_size'] == 1000
         assert client.config['sleep_time'] == 5
-        assert client.config['timeout'] == 60
+        assert client.config['timeout'] == 180
         assert 'persName' in client.config['coordinates']
         mock_configure_logging.assert_called_once()