Jukebox/jukebox-software
10
0
Fork 0
Code Issues 3 Pull Requests Packages Projects Releases 2 Wiki Activity

Create main runner app, with async multithreading

Browse Source
This commit is contained in:
Cole Deck
2024-01-17 16:06:15 -06:00
parent 33671683ea
commit 01526524d4
8 changed files with 263800 additions and 55 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
read_datasheet.py
View File

@@ -8,13 +8,14 @@ import numpy as np
from PIL import Image
import io
import json
from util import fprint
def parse(filename, output_dir):
# Extract table data
tables = camelot.read_pdf(filename, pages="1-end", flavor='lattice', backend="poppler", split_text=False, line_scale=100, process_background=True, resolution=600, interations=1, layout_kwargs={'detect_vertical': False, 'char_margin': 0.5}, shift_text=['r', 't'])
#print("Total tables extracted:", tables.n)
#fprint("Total tables extracted:", tables.n)
n = 0
pagenum = 0
reader = PdfReader(filename)
@@ -27,10 +28,10 @@ def parse(filename, output_dir):
table.df.replace(np.nan, '', inplace=True)
if not table.df.empty:
#print("\nTable " + str(n))
#fprint("\nTable " + str(n))
# Extract table names
table_start = table.cells[0][0].lt[1] # Read top-left cell's top-left coordinate
#print(table_start)
#fprint(table_start)
ymin = table_start
ymax = table_start + 10
if pagenum != table.page - 1:
@@ -46,20 +47,20 @@ def parse(filename, output_dir):
text_body = "".join(parts).strip('\n')
if len(text_body) == 0:
text_body = str(n)
#print(text_body)
#fprint(text_body)
table_list[text_body] = table.df
#table.to_html("table" + str(n) + ".html")
#print(table.df)
#fprint(table.df)
#camelot.plot(table, kind='grid').savefig("test" + str(n) + ".png")
n=n+1
#camelot.plot(tables[0], kind='grid').savefig("test.png")
#tables.export(output_dir + '/techdata.json', f='json')
# print(table_list)
# fprint(table_list)
# Extract Basic details - part name & description, image, etc
reader = PdfReader(filename)
@@ -68,7 +69,7 @@ def parse(filename, output_dir):
skip = False
for image_file_object in page.images:
if image_file_object.name == "img0.png" and skip == False:
#print(Image.open(io.BytesIO(image_file_object.data)).mode)
#fprint(Image.open(io.BytesIO(image_file_object.data)).mode)
if Image.open(io.BytesIO(image_file_object.data)).mode == "P":
skip = True
continue
@@ -137,20 +138,20 @@ def parse(filename, output_dir):
# multi-page table check
if table_name.isdigit() and len(tables) > 1:
print(table_name)
print(previous_table)
fprint(table_name)
fprint(previous_table)
main_key = previous_table
cont_key = table_name
print(tables)
fprint(tables)
if vertical == False:
main_keys = list(tables[main_key].keys())
for i, (cont_key, cont_values) in enumerate(tables[cont_key].items()):
if i < len(main_keys):
print(tables[main_key][main_keys[i]])
fprint(tables[main_key][main_keys[i]])
tables[main_key][main_keys[i]] = (tables[main_key][main_keys[i]] + (cont_key,) + cont_values)
del tables[table_name]
@@ -163,7 +164,7 @@ def parse(filename, output_dir):
previous_table = table_name
print(tables)
fprint(tables)
with open(output_dir + "/tables.json", 'w') as json_file:
json.dump(tables, json_file)