fix tablib error

liiatools/common/stream_filters.py

@@ -1,11 +1,12 @@
 import logging
 import xmlschema
 import tablib
+import pandas as pd
 import xml.etree.ElementTree as ET
 from io import BytesIO, StringIO
 from typing import Iterable, Union, Any, Dict, List
 from pathlib import Path
-from tablib import import_book, import_set
+from tablib import import_book, import_set, UnsupportedFormat
 
 from sfdata_stream_parser import events, collectors
 from sfdata_stream_parser.checks import type_check
@@ -31,6 +32,17 @@
 logger = logging.getLogger(__name__)
 
 
+def _import_set_workaround(data):
+    """
+    Workaround for a bug in tablib that causes it to fail to import
+    sets of data.
+    """
+    try:
+        return import_set(data)
+    except UnsupportedFormat:
+        return pd.read_csv(data)
+
+
 def tablib_parse(source: FileLocator):
     """
     Parse any of the tabular formats supported by TabLib
@@ -58,7 +70,7 @@ def tablib_parse(source: FileLocator):
         pass
 
     try:
-        dataset = import_set(data)
+        dataset = _import_set_workaround(data)
         logger.debug("Opened %s as a sheet", filename)
         return tablib_to_stream(dataset, filename=filename)
     except Exception as e:
@@ -86,6 +98,24 @@ def _tablib_dataset_to_stream(dataset: tablib.Dataset, **kwargs):
     yield events.EndContainer()
 
 
+def _pandas_dataframe_to_stream(dataset: pd.DataFrame, **kwargs):
+    params = {k: v for k, v in kwargs.items() if v is not None}
+    yield events.StartContainer(**params)
+    yield events.StartTable(headers=dataset.columns.tolist())
+    for r_ix, row in enumerate(dataset.itertuples(index=False)):
+        yield events.StartRow()
+        for c_ix, cell in enumerate(row[0:]):
+            yield events.Cell(
+                r_ix=r_ix,
+                c_ix=c_ix,
+                header=dataset.columns.tolist()[c_ix],
+                cell=cell,
+            )
+        yield events.EndRow()
+    yield events.EndTable()
+    yield events.EndContainer()
+
+
 def tablib_to_stream(
     data: Union[tablib.Dataset, tablib.Databook], filename: str = None
 ):
@@ -105,6 +135,9 @@ def tablib_to_stream(
                 sheet, filename=filename, sheetname=sheet.title
             )
 
+    elif isinstance(data, pd.DataFrame):
+        yield from _pandas_dataframe_to_stream(data, filename=filename)
+
 
 def inherit_property(stream, prop_name: Union[str, Iterable[str]], override=False):
     """

liiatools/school_census_pipeline/pipeline.py

@@ -25,7 +25,6 @@ def process_file(
     pipeline_config: PipelineConfig,
     la_code: str,
 ) -> ProcessResult:
-    print(f"{file_locator=}")
     errors = ErrorContainer()
     year = pl.discover_year(file_locator)
     if year is None: