From 423233da6f6899b57fd15c422b70631dd9b5ea79 Mon Sep 17 00:00:00 2001
From: Constantin Pape <c.pape@gmx.net>
Date: Fri, 21 Jun 2019 12:47:11 +0200
Subject: [PATCH] Implemnt file renaming WIP
---
make_data_sources.py | 51 ++++++++++++++++++++++++++++++++++++++
scripts/__init__.py | 1 +
scripts/files/__init__.py | 1 +
scripts/files/xml_utils.py | 14 +++++++++++
scripts/sources.py | 25 +++++++++++++++++++
5 files changed, 92 insertions(+)
create mode 100644 make_data_sources.py
create mode 100644 scripts/sources.py
diff --git a/make_data_sources.py b/make_data_sources.py
new file mode 100644
index 0000000..1c62b54
--- /dev/null
+++ b/make_data_sources.py
@@ -0,0 +1,51 @@
+import os
+import glob
+from .scripts.files import get_h5_path_from_xml, copy_xml_with_newpath
+
+
+def copy_xmls_and_symlink_h5(name_dict, src_folder, trgt_folder):
+ for n1, n2 in name_dict.items():
+ src_xml = os.path.join(src_folder, n1)
+ trgt_xml = os.path.join(trgt_folder, n2)
+
+ # we make a softlink from src to target h5
+ # NOTE eventually, we want to copy all the data, but
+ # for now, we use softlinks in order to keep the current
+ # version of the platy browser working
+ src_h5 = get_h5_path_from_xml(src_xml, return_absolute_path=True)
+ trgt_h5 = os.path.splitext(n2)[0] + '.h5'
+ os.symlink(src_h5, trgt_h5)
+
+ copy_xml_with_newpath(src_xml, trgt_xml, trgt_h5)
+
+
+def make_initial_data_sources():
+ old_folder = '/g/arendt/EM_6dpf_segmentation/EM-Prospr'
+ raw_folder = './data/rawdata'
+
+ # copy the sbem data
+ sbem_prefix = 'sbem-6dpf-1-whole'
+ name_dict = {'em-raw-', 'em-raw',
+ 'em-segmented-', ''}
+ name_dict = {k: '%s-%s' % (sbem_prefix, v)
+ for k, v in name_dict.items()}
+ copy_xmls_and_symlink_h5(name_dict, old_folder, raw_folder)
+
+ # copy the prospr meds
+ prospr_prefix = 'prospr-6dpf-1-whole'
+ prospr_names = glob.glob(os.path.join(old_folder, "*-MED*"))
+ prospr_names = [os.path.split(f) for f in prospr_names]
+ name_dict = {n: '%s-%s' % (prospr_prefix, n) for n in prospr_names}
+ copy_xmls_and_symlink_h5(name_dict, old_folder, raw_folder)
+
+ # copy the fibsem data
+ fib_prefix = 'fibsem-6dpf-1-parapod'
+ name_dict = {'em-raw-', 'em-raw',
+ 'em-segmented-', ''}
+ name_dict = {k: '%s-%s' % (fib_prefix, v)
+ for k, v in name_dict.items()}
+ copy_xmls_and_symlink_h5(name_dict, old_folder, raw_folder)
+
+
+if __name__ == '__main__':
+ make_initial_data_sources()
diff --git a/scripts/__init__.py b/scripts/__init__.py
index e69de29..fdf8054 100644
--- a/scripts/__init__.py
+++ b/scripts/__init__.py
@@ -0,0 +1 @@
+from .sources import get_name_prefixes
diff --git a/scripts/files/__init__.py b/scripts/files/__init__.py
index 111cf66..bc36dfb 100644
--- a/scripts/files/__init__.py
+++ b/scripts/files/__init__.py
@@ -2,4 +2,5 @@ from .copy_helper import copy_tables, copy_segmentation, copy_static_files
from .copy_helper import copy_files_with_pattern
from .folders import make_folder_structure
from .xml_utils import get_h5_path_from_xml, copy_xml_with_abspath, write_simple_xml
+from .xml_utils import copy_xml_with_newpath
from .bdv_server import make_bdv_server_file
diff --git a/scripts/files/xml_utils.py b/scripts/files/xml_utils.py
index e297343..7021d71 100644
--- a/scripts/files/xml_utils.py
+++ b/scripts/files/xml_utils.py
@@ -58,6 +58,20 @@ def copy_xml_with_abspath(xml_in, xml_out):
tree.write(xml_out)
+def copy_xml_with_newpath(xml_in, xml_out, h5path):
+ # get the h5 path from the xml
+ et_root = ET.parse(xml_in).getroot()
+ et = et_root[1]
+ et = et[0]
+ et = et[0]
+ # write new xml with the new path
+ et.text = h5path
+ et.set('type', 'absolute')
+ indent_xml(et_root)
+ tree = ET.ElementTree(et_root)
+ tree.write(xml_out)
+
+
def write_simple_xml(xml_path, h5_path, path_type='absolute'):
# write top-level data
root = ET.Element('SpimData')
diff --git a/scripts/sources.py b/scripts/sources.py
new file mode 100644
index 0000000..9e736b1
--- /dev/null
+++ b/scripts/sources.py
@@ -0,0 +1,25 @@
+# this folder contains information about the current data sources, see
+# https://git.embl.de/tischer/platy-browser-tables/blob/dev/README.md#file-naming
+
+# TODO maybe store this as exteral file in json
+# list of the current data sources
+SOURCES = [{'modality': 'sbem', 'stage': '6dpf', 'id': '1', 'region': 'whole'},
+ {'modality': 'prospr', 'stage': '6dpf', 'id': '1', 'region': 'whole'},
+ {'modality': 'fibsem', 'stage': '6dpf', 'id': '1', 'region': 'parapod'}]
+
+
+def get_sources():
+ return SOURCES
+
+
+def source_to_prefix(source):
+ return '%s-%s-%s-%s' % (source['modality'],
+ source['stage'],
+ source['id'],
+ source['region'])
+
+
+def get_name_prefixes():
+ sources = get_sources()
+ prefixes = [source_to_prefix(source) for source in sources]
+ return prefixes
--
GitLab