summaryrefslogtreecommitdiff
path: root/animism-align/cli/app
diff options
context:
space:
mode:
Diffstat (limited to 'animism-align/cli/app')
-rw-r--r--animism-align/cli/app/settings/app_cfg.py10
-rw-r--r--animism-align/cli/app/utils/file_utils.py12
2 files changed, 17 insertions, 5 deletions
diff --git a/animism-align/cli/app/settings/app_cfg.py b/animism-align/cli/app/settings/app_cfg.py
index 492b5ed..69a6adc 100644
--- a/animism-align/cli/app/settings/app_cfg.py
+++ b/animism-align/cli/app/settings/app_cfg.py
@@ -106,3 +106,13 @@ ROMAN_NUMERALS = [
'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X',
'XI', 'XII', 'XIII', 'XIV', 'XV', 'XVI', 'XVII', 'XVIII', 'XIX', 'XX',
]
+TEXT_ANNOTATION_TYPES = [
+ 'section_heading',
+ # 'heading_text',
+ 'sentence',
+ # 'paragraph_end',
+ # 'pullquote_credit',
+ 'footnote',
+ # 'text_plate',
+ # 'subtitle',
+]
diff --git a/animism-align/cli/app/utils/file_utils.py b/animism-align/cli/app/utils/file_utils.py
index 1d19fd6..c61e779 100644
--- a/animism-align/cli/app/utils/file_utils.py
+++ b/animism-align/cli/app/utils/file_utils.py
@@ -195,12 +195,14 @@ def load_yaml(fp_in):
cfg = yaml.load(fp, Loader=yaml.Loader)
return cfg
-def load_text(fp_in):
+def load_text(fp_in, split=True):
"""Load a text file into an array
:param fp_in: (str) filepath
"""
with open(fp_in, 'rt') as fp:
- lines = fp.read().rstrip('\n').split('\n')
+ lines = fp.read().rstrip('\n')
+ if split:
+ lines = lines.split('\n')
return lines
def load_line_lookup(fp_in):
@@ -264,16 +266,16 @@ def write_pickle(data, fp_out, ensure_path=True):
pickle.dump(data, fp)
-def write_json(data, fp_out, minify=True, ensure_path=True, sort_keys=True, verbose=False):
+def write_json(data, fp_out, minify=True, ensure_path=True, sort_keys=True, verbose=False, default=None):
"""
"""
if ensure_path:
mkdirs(fp_out)
with open(fp_out, 'w') as fp:
if minify:
- json.dump(data, fp, separators=(',',':'), sort_keys=sort_keys)
+ json.dump(data, fp, separators=(',',':'), sort_keys=sort_keys, default=default)
else:
- json.dump(data, fp, indent=2, sort_keys=sort_keys)
+ json.dump(data, fp, indent=2, sort_keys=sort_keys, default=default)
if verbose:
log.info('Wrote JSON: {}'.format(fp_out))