Cleaned up notebook

.gitignore

@@ -16,3 +16,7 @@ output.mp4
 output.log
 # images
 *.png
+# Built app
+build
+# Generated label images
+labels

Dockerfile

@@ -1,7 +1,9 @@
-FROM python:latest
+FROM python:3.11-slim
 
-RUN apt-get update && apt-get install -y libgl1-mesa-glx ghostscript && apt-get clean && rm -rf /var/lib/apt/lists
-COPY . .
+# Get runtime dependencies
+# glx for OpenCV, ghostscript for datasheet PDF rendering, zbar for barcode scanning, git for cloning repos
+RUN apt-get update && apt-get install -y libgl1-mesa-glx ghostscript libzbar0 git && apt-get clean && rm -rf /var/lib/apt/lists
+COPY *.py *.yml *.sh *.txt *.html static templates ./
 #COPY config-server.yml config.yml
 RUN pip3 install -r requirements.txt
 

get_specs.py

@@ -27,6 +27,7 @@ def check_internet(url='https://belden.com', timeout=5):
 
 
 def query_search(partnum, source):
+    fprint("Searching for " + partnum)
     if source == "Belden":
         token_url = "https://www.belden.com/coveo/rest/token?t=" + str(int(time.time()))
         with requests.get(token_url) as r:
@@ -52,15 +53,15 @@ def query_search(partnum, source):
                 a = json.loads(a)
                 idx = -1
                 name = ""
-                for partid in range(len(a["results"])):
+                for partid in range(len(a["results"])-1, -1, -1):
                     name = a["results"][partid]["title"]
                     if name != partnum:
                         if name.find(partnum) >= 0:
                             idx = partid
-                            break
+                            #break
                         elif partnum.find(name) >= 0:
                             idx = partid
-                            break
+                            #break
                         
                     else:
                         idx = partid
@@ -69,7 +70,7 @@ def query_search(partnum, source):
                 if idx < 0:
                     fprint("Could not find part in API: " + partnum)
                     return False
-                fprint("Search result found: result " + str(idx) + ", for ID " + name)
+                #fprint("Search result found: result " + str(idx) + ", for ID " + name)
                 #urlname = a["results"][0]["raw"]["catalogitemurlname"]
                 img = a["results"][idx]["raw"]["catalogitemimageurl"]
                 img = img[0:img.index("?")]
@@ -92,7 +93,7 @@ def query_search(partnum, source):
                 #print(out)
                 return out
         except:
-            print("falied to search with API. Falling back to datasheet lookup.")
+            print("Failed to search with API. Falling back to datasheet lookup.")
             return False
     
 
@@ -112,12 +113,14 @@ def query_search(partnum, source):
         r = requests.get(url=alphaurl)
         data = r.json()
         output = dict()
-        #print(data)
+        #print(data["Results"])
+        
         try:
             if data["Count"] > 0:
                 #print(data["Results"][0]["Url"])
                 for result in data["Results"]:
-                    if result["Url"].split("/")[-1] == partnum:
+                    #print(result["Url"])
+                    if result["Url"].split("/")[-1] == partnum.replace("-", "").replace("/", "_"):
                         #print(partnum)
                         #print(result["Html"])
                         try:
@@ -133,14 +136,18 @@ def query_search(partnum, source):
                         dsidx = result["Html"].index("<a href=\"/disteAPI/") + 9
                         dsidx2 = result["Html"].index(partnum, dsidx) + len(partnum)
                         output["datasheet"] = "https://www.alphawire.com" + result["Html"][dsidx:dsidx2]
-                        output["partnum"] = partnum
-                        #"test".index()
+                        output["partnum"] = partnum.replace("/", "_") #.replace("-", "").replace("/", "_")
+                        #
+                        # "test".index()
                         #print(output)
                         return output
 
 
         except:
+            print("Failed to search with API. Falling back to datasheet lookup.")
             return False
+        
+        print("Failed to search with API. Falling back to datasheet lookup.")
         return False
 
 
@@ -150,14 +157,19 @@ def touch(path):
 
 
 
-def get_multi(partnums, delay=0.25):
+def get_multi(partnums, delay=0.25, dir="cables/", cache=True):
     with alive_bar(len(partnums) * 2, dual_line=True, calibrate=30, bar="classic2", spinner="classic") as bar:
-
-        def _try_download_datasheet(partnum, output_dir): # Guess datasheet URL
+        failed = list()
+        actualpartnums = list()
+        def _try_download_datasheet(partnum, output_dir, dstype): # Guess datasheet URL
             global bartext
 
+            if dstype == "Belden":
                 sanitized_name = partnum.replace(" ", "")
                 url = "https://catalog.belden.com/techdata/EN/" + sanitized_name + "_techdata.pdf"
+            elif dstype == "Alphawire":
+                # Alphawire Datasheet URLs do not use a sanitized part number (but product pages do)
+                url = "https://www.alphawire.com/disteAPI/SpecPDF/DownloadProductSpecPdf?productPartNumber=" + partnum
             #fprint(url)
             try:
                 with requests.get(url, stream=True) as r:
@@ -167,8 +179,9 @@ def get_multi(partnums, delay=0.25):
                     if r.status_code == 404:
                         return False
                     os.makedirs(output_dir, exist_ok=True)
+                    bartext = ""
                     with open(output_dir + "/datasheet.pdf", 'wb') as f:
-                        for chunk in r.iter_content(chunk_size=131072): 
+                        for chunk in r.iter_content(chunk_size=65536): 
                             # If you have chunk encoded response uncomment if
                             # and set chunk_size parameter to None.
                             #if chunk: 
@@ -195,8 +208,9 @@ def get_multi(partnums, delay=0.25):
                     if r.status_code == 404:
                         return False
                     os.makedirs(output_dir, exist_ok=True)
+                    bartext = ""
                     with open(output_dir + "/datasheet.pdf", 'wb') as f:
-                        for chunk in r.iter_content(chunk_size=131072): 
+                        for chunk in r.iter_content(chunk_size=65536): 
                             # If you have chunk encoded response uncomment if
                             # and set chunk_size parameter to None.
                             #if chunk: 
@@ -221,8 +235,9 @@ def get_multi(partnums, delay=0.25):
                     if r.status_code == 404:
                         return False
                     os.makedirs(output_dir, exist_ok=True)
+                    bartext = ""
                     with open(output_dir + "/part-hires." + url.split(".")[-1], 'wb') as f:
-                        for chunk in r.iter_content(chunk_size=131072): 
+                        for chunk in r.iter_content(chunk_size=65536): 
                             # If you have chunk encoded response uncomment if
                             # and set chunk_size parameter to None.
                             #if chunk: 
@@ -245,8 +260,9 @@ def get_multi(partnums, delay=0.25):
                 fprint("Parsing Datasheet contents of " + partnum)
                 bar.text = "Parsing Datasheet contents of " + partnum + ".pdf..."
             
-                read_datasheet.parse(path, output_dir, partnum, dstype)
+                out = read_datasheet.parse(path, output_dir, partnum, dstype)
                 bar(skipped=False) 
+                return out
             else:
                 fprint("Datasheet already parsed for " + partnum)
                 bar.text = "Datasheet already parsed for " + partnum + ".pdf"
@@ -258,57 +274,73 @@ def get_multi(partnums, delay=0.25):
             bar(skipped=False)
             fprint("Parsing Datasheet contents of " + partnum)
             bar.text = "Parsing Datasheet contents of " + partnum + ".pdf..."
-            read_datasheet.parse(path, output_dir, partnum, dstype)
+            out = read_datasheet.parse(path, output_dir, partnum, dstype)
             bar(skipped=False)
+            return out
 
         def run_search(partnum):
-            output_dir = "cables/" + partnum
+            oldpartnum = partnum
+            if dstype == "Alphawire":
+                # For alphawire, sanitize the part number for only the final result check, because their API is very wierd
+                # For the actual search, it must be un-sanitized
+                partnum = partnum.replace("/","_")
+            output_dir = dir + partnum
             path = output_dir + "/datasheet.pdf"
             bartext = "Downloading files for part " + partnum
             bar.text = bartext
-            #
-            if (not os.path.exists(output_dir + "/found_part_hires")) or not (os.path.exists(path) and os.path.getsize(path) > 1):
+            partnum = oldpartnum.replace("_","/")
+            returnval = [partnum, dstype, False, False]
+            if (not os.path.exists(output_dir + "/found_part_hires")) or not (os.path.exists(path) and os.path.getsize(path) > 1) or not cache:
                 # Use query
                 search_result = query_search(partnum, dstype)
                 # Try to use belden.com search
                 if search_result is not False:
                     # Download high resolution part image if available and needed
+                    #oldpartnum = partnum
                     partnum = search_result["partnum"]
-                    output_dir = "cables/" + partnum
+                    returnval = [partnum, dstype, False, False]
+                    output_dir = dir + partnum
                     path = output_dir + "/datasheet.pdf"
                     bartext = "Downloading files for part " + partnum
                     bar.text = bartext
 
-                    if not os.path.exists(output_dir + "/found_part_hires"):
+                    if not os.path.exists(output_dir + "/found_part_hires") or not cache:
                         if _download_image(search_result["image"], output_dir):
                             fprint("Downloaded hi-res part image for " + partnum)
+                            returnval = [partnum, dstype, True, False]
                         touch(output_dir + "/found_part_hires")
                     else:
                         fprint("Using cached hi-res part image for " + partnum)
 
                     # Download datasheet from provided URL if needed
-                    if os.path.exists(path) and os.path.getsize(path) > 1:
-                        __use_cached_datasheet(partnum, path, output_dir, dstype)
+                    if os.path.exists(path) and os.path.getsize(path) > 1 and cache:
+                        out = __use_cached_datasheet(partnum, path, output_dir, dstype)
+                        returnval = [partnum, dstype, True, out]
 
                     elif _download_datasheet(search_result["datasheet"], output_dir) is not False:
-                        __downloaded_datasheet(partnum, path, output_dir, dstype)
+                        out = __downloaded_datasheet(partnum, path, output_dir, dstype)
+                        returnval = [partnum, dstype, True, out]
                 
-                elif os.path.exists(path) and os.path.getsize(path) > 1:
-                    __use_cached_datasheet(partnum, path, output_dir, dstype)
+                elif os.path.exists(path) and os.path.getsize(path) > 1 and cache:
+                    out = __use_cached_datasheet(partnum, path, output_dir, dstype)
+                    returnval = [partnum, dstype, True, out]
                 
                 # If search fails, and we don't already have the datasheet, guess datasheet URL and skip the hires image download
-                elif _try_download_datasheet(partnum, output_dir) is not False:
-                    __downloaded_datasheet(partnum, path, output_dir, dstype)
+                elif _try_download_datasheet(partnum, output_dir, dstype) is not False:
+                    out = __downloaded_datasheet(partnum, path, output_dir, dstype)
+                    returnval = [partnum, dstype, False, out]
 
                 # Failed to download with search or guess :(
                 else: 
                     return False
-                return True
+                actualpartnums.append(returnval)
+                return returnval
 
             # We already have a hi-res image and the datasheet - perfect!
             else:
                 fprint("Using cached hi-res part image for " + partnum)
-                __use_cached_datasheet(partnum, path, output_dir, dstype)
+                out = __use_cached_datasheet(partnum, path, output_dir, dstype)
+                returnval = [partnum, dstype, False, out]
                 return True
     
         for fullpartnum in partnums:
@@ -339,7 +371,7 @@ def get_multi(partnums, delay=0.25):
                 if not success:
                     fprint("Failed to download datasheet for part " + partnum)
                     bar.text = "Failed to download datasheet for part " + partnum
-                    failed.append(partnum)
+                    failed.append((partnum, dstype))
                     bar(skipped=True)
                     bar(skipped=True)
             time.sleep(delay)
@@ -347,10 +379,10 @@ def get_multi(partnums, delay=0.25):
     if len(failed) > 0:
         fprint("Failed to download:")
         for partnum in failed:
-            fprint(partnum)
-        return False # Go to manual review upload page
+            fprint(partnum[1] + " " + partnum[0])
+        return False, actualpartnums # Go to manual review upload page
     else:
-        return True # All cables downloaded; we are good to go
+        return True, actualpartnums # All cables downloaded; we are good to go
 
 
 
@@ -377,7 +409,7 @@ if __name__ == "__main__":
     "AW3050",
     "AW6714",
     "AW1172C",
-    "AW2211/4",
+    "AWFIT-221-1/4",
 
     "BLTF-1LF-006-RS5N",
     "BLTF-SD9-006-RI5N",
@@ -421,8 +453,8 @@ if __name__ == "__main__":
     "BLC6D1100007"
 
     ]
-    #query_search("86104CY", "Alphawire")
-    get_multi(partnums, 0.25)
+    print(query_search("74002", "Belden"))
+    #get_multi(partnums, 0.25)
     #query_search("10GXS13", "Belden")
 
 

label_generator.py

@@ -0,0 +1,100 @@
+#!/usr/bin/env python3
+
+
+
+from get_specs import get_multi
+import sys
+import uuid
+import os
+import signal
+from PIL import Image
+from label_image import generate_code
+
+
+def input_cable():
+    print("")
+    print("Use the full part number. Spaces, special characters are allowed. Do not specify the brand.")
+    print("")
+    print("Please enter a part number and press enter:")
+    inputnum = input("").strip()
+    if len(inputnum) < 2:
+        killall_signal(0, 0)
+    print("Input part number:", inputnum)
+    print("Searching databases for cables...")
+    # Search both AW and BL sites
+    status, output = get_multi(["BL"+inputnum, "AW"+inputnum], delay=0.1, dir="temp/" + str(uuid.uuid4()) + "/", cache=False)
+    print("")
+    if len(output) > 1:
+        for i in output:
+            print(i[1], i[0])
+        print("Multiple brands with the same part number! Please type \"b\" for the Belden part number or \"a\" for the Alphawire cable")
+        inputbrand = input()
+        if inputbrand == "b":
+            output = [output[0]]
+        elif inputbrand == "a":
+            output = [output[1]]
+    elif len(output) == 0:
+        print("No results found for part number", inputnum + ". Please try again with a different part number.")
+        return
+    
+    output = output[0]
+    print("")
+    if output[2] and output[3]:
+        print("Cable result found -",output[1], output[0], "with high-quality image and full specs")
+    elif output[2]:
+        print("Cable result found -",output[1], output[0], "with high-quality image and no specs")
+    elif output[3]:
+        print("Cable result found -",output[1], output[0], "with no/low quality image and full specs")
+    else:
+        print("Cable result found -",output[1], output[0], "with no/low quality image and no specs")
+    print("")
+    if not output[3]:
+        print("Unable to decode cable specs. Please try again with a different part number.")
+        return False
+    else:
+        print("")
+        print("*** Cable details confirmed. Creating label...")
+        print("")
+        img = None
+        imgstr = ""
+        if output[1] == "Belden":
+            imgstr = "BL"
+        elif output[1] == "Alphawire":
+            imgstr = "AW"
+        img = generate_code(imgstr + output[0])
+        os.makedirs("labels", exist_ok=True)
+        img.save("labels/" + imgstr + output[0] + ".png")
+
+def delete_folder(path):
+    # Check if the path is a directory
+    if not os.path.isdir(path):
+        return
+    
+    # List all files and directories in the path
+    for filename in os.listdir(path):
+        file_path = os.path.join(path, filename)
+        # If it's a directory, recursively call this function
+        if os.path.isdir(file_path):
+            delete_folder(file_path)
+        else:
+            # If it's a file, remove it
+            os.remove(file_path)
+    
+    # After removing all contents, remove the directory itself
+    os.rmdir(path)
+
+def killall_signal(a,b):
+    delete_folder("temp")
+    os.kill(os.getpid(), 9) # dirty kill of self
+
+if __name__ == "__main__":
+    
+    signal.signal(signal.SIGINT, killall_signal)
+    signal.signal(signal.SIGTERM, killall_signal)
+    print("Welcome to the Jukebox cable utility. This tool will allow you to verify Belden & Alphawire cable part numbers and create labels for samples in the Jukebox.")
+    print("This tool requires internet access to download cable specifications and verify part numbers.")
+    #print("Use Ctrl+C to exit.")
+    while True:
+        delete_folder("temp")
+        input_cable()
+    

label_image.py

@@ -0,0 +1,238 @@
+#!/usr/bin/env python3
+from util import fprint
+
+from PIL import Image
+from PIL import ImageDraw
+#import cv2
+import numpy as np
+#import math
+
+
+
+# Copied from http://en.wikipedia.org/wiki/Code_128
+# Value Weights 128A    128B    128C
+CODE128_CHART = """
+0       212222  space   space   00
+1       222122  !       !       01
+2       222221  "       "       02
+3       121223  #       #       03
+4       121322  $       $       04
+5       131222  %       %       05
+6       122213  &       &       06
+7       122312  '       '       07
+8       132212  (       (       08
+9       221213  )       )       09
+10      221312  *       *       10
+11      231212  +       +       11
+12      112232  ,       ,       12
+13      122132  -       -       13
+14      122231  .       .       14
+15      113222  /       /       15
+16      123122  0       0       16
+17      123221  1       1       17
+18      223211  2       2       18
+19      221132  3       3       19
+20      221231  4       4       20
+21      213212  5       5       21
+22      223112  6       6       22
+23      312131  7       7       23
+24      311222  8       8       24
+25      321122  9       9       25
+26      321221  :       :       26
+27      312212  ;       ;       27
+28      322112  <       <       28
+29      322211  =       =       29
+30      212123  >       >       30
+31      212321  ?       ?       31
+32      232121  @       @       32
+33      111323  A       A       33
+34      131123  B       B       34
+35      131321  C       C       35
+36      112313  D       D       36
+37      132113  E       E       37
+38      132311  F       F       38
+39      211313  G       G       39
+40      231113  H       H       40
+41      231311  I       I       41
+42      112133  J       J       42
+43      112331  K       K       43
+44      132131  L       L       44
+45      113123  M       M       45
+46      113321  N       N       46
+47      133121  O       O       47
+48      313121  P       P       48
+49      211331  Q       Q       49
+50      231131  R       R       50
+51      213113  S       S       51
+52      213311  T       T       52
+53      213131  U       U       53
+54      311123  V       V       54
+55      311321  W       W       55
+56      331121  X       X       56
+57      312113  Y       Y       57
+58      312311  Z       Z       58
+59      332111  [       [       59
+60      314111  \       \       60
+61      221411  ]       ]       61
+62      431111  ^       ^       62
+63      111224  _       _       63
+64      111422  NUL     `       64
+65      121124  SOH     a       65
+66      121421  STX     b       66
+67      141122  ETX     c       67
+68      141221  EOT     d       68
+69      112214  ENQ     e       69
+70      112412  ACK     f       70
+71      122114  BEL     g       71
+72      122411  BS      h       72
+73      142112  HT      i       73
+74      142211  LF      j       74
+75      241211  VT      k       75
+76      221114  FF      l       76
+77      413111  CR      m       77
+78      241112  SO      n       78
+79      134111  SI      o       79
+80      111242  DLE     p       80
+81      121142  DC1     q       81
+82      121241  DC2     r       82
+83      114212  DC3     s       83
+84      124112  DC4     t       84
+85      124211  NAK     u       85
+86      411212  SYN     v       86
+87      421112  ETB     w       87
+88      421211  CAN     x       88
+89      212141  EM      y       89
+90      214121  SUB     z       90
+91      412121  ESC     {       91
+92      111143  FS      |       92
+93      111341  GS      }       93
+94      131141  RS      ~       94
+95      114113  US      DEL     95
+96      114311  FNC3    FNC3    96
+97      411113  FNC2    FNC2    97
+98      411311  ShiftB  ShiftA  98
+99      113141  CodeC   CodeC   99
+100     114131  CodeB   FNC4    CodeB
+101     311141  FNC4    CodeA   CodeA
+102     411131  FNC1    FNC1    FNC1
+103     211412  StartA  StartA  StartA
+104     211214  StartB  StartB  StartB
+105     211232  StartC  StartC  StartC
+106     2331112 Stop    Stop    Stop
+""".split()
+
+VALUES   = [int(value) for value in CODE128_CHART[0::5]]
+WEIGHTS  = dict(zip(VALUES, CODE128_CHART[1::5]))
+CODE128A = dict(zip(CODE128_CHART[2::5], VALUES))
+CODE128B = dict(zip(CODE128_CHART[3::5], VALUES))
+CODE128C = dict(zip(CODE128_CHART[4::5], VALUES))
+
+for charset in (CODE128A, CODE128B):
+    charset[' '] = charset.pop('space')
+
+
+
+def generate_code(data, show=False, check=False):
+
+    img = code128_image(data)
+    if show:
+        img.show()
+    #img.show()
+    #print(data)
+    
+    if(check):
+        from pyzbar.pyzbar import decode
+        from pyzbar.pyzbar import ZBarSymbol
+        print(decode(img, symbols=[ZBarSymbol.CODE128])[0].data.decode('ascii'))
+
+    #if(decode(img, symbols=[ZBarSymbol.CODE128])[0].data.decode('ascii') == data):
+    #    return True
+    #else:
+    #    return False
+    return img
+
+def code128_format(data):
+    """
+    Generate an optimal barcode from ASCII text
+    """
+    text     = str(data)
+    pos      = 0
+    length   = len(text)
+    
+    # Start Code
+    if text[:2].isdigit():
+        charset = CODE128C
+        codes   = [charset['StartC']]
+    else:
+        charset = CODE128B
+        codes   = [charset['StartB']]
+    
+    # Data
+    while pos < length:
+        if charset is CODE128C:
+            if text[pos:pos+2].isdigit() and length - pos > 1:
+                # Encode Code C two characters at a time
+                codes.append(int(text[pos:pos+2]))
+                pos += 2
+            else:
+                # Switch to Code B
+                codes.append(charset['CodeB'])
+                charset = CODE128B
+        elif text[pos:pos+4].isdigit() and length - pos >= 4:
+            # Switch to Code C
+            codes.append(charset['CodeC'])
+            charset = CODE128C
+        else:
+            # Encode Code B one character at a time
+            codes.append(charset[text[pos]])
+            pos += 1
+    
+    # Checksum
+    checksum = 0
+    for weight, code in enumerate(codes):
+        checksum += max(weight, 1) * code
+    codes.append(checksum % 103)
+    
+    # Stop Code
+    codes.append(charset['Stop'])
+    return codes
+
+def code128_image(data, height=100, thickness=3, quiet_zone=False):
+    if not data[-1] == CODE128B['Stop']:
+        data = code128_format(data)
+        
+    
+    barcode_widths = []
+    for code in data:
+        for weight in WEIGHTS[code]:
+            barcode_widths.append(int(weight) * thickness)
+    width = sum(barcode_widths)
+    x = 0
+    
+
+    if quiet_zone:
+        width += 20 * thickness
+        x = 10 * thickness
+
+    # Monochrome Image
+    img  = Image.new('RGB', (int(width * 10), int(width * 10)), 'white')
+    draw = ImageDraw.Draw(img)
+    draw_bar = True
+    for bwidth in barcode_widths:
+        bwidth *= 4
+        if draw_bar:
+            draw.rectangle(((x + int(width * 3), width*6.25), (x + int(width * 3) + bwidth - 1, width*7)), fill='black')
+        draw_bar = not draw_bar
+        x += bwidth
+
+    #draw.arc(((width - width/5, width - width/5), (width*9 + width/5, width*9 + width/5)),0,360,fill='blue', width = int(width/8))
+    draw.arc(((width+int(width / 1.4), width+int(width / 1.4)), (width*9-int(width / 1.4), width*9-int(width / 1.4))),0,360,fill='blue', width = int(width/8))
+    return img
+
+if __name__ == "__main__":
+    #print(generate_code("BL10GXS13"))
+    #print(generate_code("BL10GXgd35j35S13"))
+    #print(generate_code("BL10GX54hS13"))
+    print(generate_code("BL10Gj34qXS13", False, False))
+    #print(generate_code("BL104w5545dp7bfwp43643534/4563G-XS13"))
+    #adjust_image(cv2.imread('test_skew.jpg'))

read_datasheet.py

@@ -1,9 +1,8 @@
 #!/usr/bin/env python3
 
-# Parse Belden catalog techdata datasheets 
+# Parse Belden (100%) & Alphawire (75%) catalog techdata datasheets 
 
 import pandas as pd
-pd.set_option('future.no_silent_downcasting', True)
 from PyPDF2 import PdfReader
 import camelot
 import numpy as np
@@ -13,55 +12,94 @@ import json
 from util import fprint
 import uuid
 from util import run_cmd
+from util import win32
 import os
+import glob
+import sys
 
 def touch(path):
     with open(path, 'a'):
         os.utime(path, None)
 
+def find_data_file(filename):
+    if getattr(sys, "frozen", False):
+        # The application is frozen
+        datadir = os.path.dirname(sys.executable)
+    else:
+        # The application is not frozen
+        # Change this bit to match where you store your data files:
+        datadir = os.path.dirname(__file__)
+    return os.path.join(datadir, filename)
+
+def extract_table_name(table_start, searchpage, reader, dstype, fallbackname):
+    if dstype == "Belden":
+        ymin = table_start
+        ymax = table_start + 10
+    elif dstype == "Alphawire":
+        ymin = table_start - 5
+        ymax = table_start + 10
+    page = reader.pages[searchpage - 1]
+    parts = []
+    def visitor_body(text, cm, tm, fontDict, fontSize):
+        y = tm[5]
+        if y > ymin and y < ymax:
+            parts.append(text)
+    page.extract_text(visitor_text=visitor_body)
+    text_body = "".join(parts).strip('\n')
+    if len(text_body) == 0:
+        text_body = str(fallbackname)
+
+    return text_body
+    #fprint(text_body)
+
 def parse(filename, output_dir, partnum, dstype):
-
+    tables = []
     # 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'])
+    try:
+        if dstype == "Belden":
+            tables = camelot.read_pdf(filename, pages="1-end", flavor='lattice', backend="ghostscript", 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'])
+        elif dstype == "Alphawire":
+            tables = camelot.read_pdf(filename, pages="1-end", flavor='lattice', backend="ghostscript", split_text=False, line_scale=50, process_background=True, resolution=600, interations=1, layout_kwargs={'detect_vertical': True, 'char_margin': 0.5}, shift_text=['l', 't'])
+    except (OSError, RuntimeError) as e: 
+        print(e)
+        if win32:
+            print("Ghostscript is not installed! Launching installer...")
+            #subprocess.run([r".\\gs10030w64.exe"])
+            os.system(r'''Powershell -Command "& { Start-Process \"''' + find_data_file("gs10030w64.exe") + r'''\" -Verb RunAs } " ''')
+            # Will return once file launched...
+            print("Once the install is completed, try again.")
+            return False
+        else:
+            print("Ghostscript is not installed. You can install it with e.g. apt install ghostscript for Debian-based systems.")
+            return False
     #fprint("Total tables extracted:", tables.n)
     n = 0
-    pagenum = 0
+    #pagenum = 0
     reader = PdfReader(filename)
     page = reader.pages[0]
     table_list = {}
-
+    table_list_raw = {}
+    pd.set_option('future.no_silent_downcasting', True)
     for table in tables:
+        #with pd.options.context("future.no_silent_downcasting", True):
         table.df.infer_objects(copy=False)
-        table.df.replace('', np.nan, inplace=True)
+        table.df = table.df.replace('', np.nan).infer_objects(copy=False)
         table.df.dropna(inplace=True, how="all")
         table.df.dropna(inplace=True, axis="columns", how="all")
-        table.df.replace(np.nan, '', inplace=True)
+        table.df = table.df.replace(np.nan, '').infer_objects(copy=False)
         
         if not table.df.empty:
             #fprint("\nTable " + str(n))
             # Extract table names
             table_start = table.cells[0][0].lt[1] # Read top-left cell's top-left coordinate
             #fprint(table_start)
-            ymin = table_start
-            ymax = table_start + 10
-            if pagenum != table.page - 1:
-                pagenum = table.page - 1
-                page = reader.pages[table.page - 1]
-            parts = []
-            def visitor_body(text, cm, tm, fontDict, fontSize):
-                y = tm[5]
-                if y > ymin and y < ymax:
-                    parts.append(text)
-
-            page.extract_text(visitor_text=visitor_body)
-            text_body = "".join(parts).strip('\n')
-            if len(text_body) == 0:
-                text_body = str(n)
-            #fprint(text_body)
             
+            text_body = extract_table_name(table_start, table.page, reader, dstype, n)
 
             table_list[text_body] = table.df
+            if dstype == "Alphawire":
+                table_list_raw[text_body] = table
+
             #table.to_html("table" + str(n) + ".html")
             
             #fprint(table.df)
@@ -71,7 +109,7 @@ def parse(filename, output_dir, partnum, dstype):
     
     #tables.export(output_dir + '/techdata.json', f='json')
 
-    # fprint(table_list)
+    #fprint(table_list)
     # Extract Basic details - part name & description, image, etc
 
     reader = PdfReader(filename)
@@ -100,24 +138,32 @@ def parse(filename, output_dir, partnum, dstype):
     tables = dict()
     torename = dict()
     previous_table = ""
+    #print(table_list.keys())
     for table_name in table_list.keys():
         # determine shape: horizontal or vertical
         table = table_list[table_name]
         rows = table.shape[0]
         cols = table.shape[1]
         vertical = None
+        #print(rows, cols, table_name)
         if rows > 2 and cols == 2:
             vertical = True
-        elif cols == 1:
+        elif cols == 1 and rows > 1:
             vertical = False
         elif rows == 1:
             vertical = True
         elif cols == 2: # and rows <= 2
             # inconsistent
+            if dstype == "Belden":
                 if table.iloc[0, 0].find(":") == len(table.iloc[0, 0]) - 1: # check if last character is ":" indicating a vertical table
                     vertical = True
                 else:
                     vertical = False
+            elif dstype == "Alphawire":
+                if table.iloc[0, 0].find(")") == 1 or table.iloc[0, 0].find(")") == 2 or table.iloc[0, 0].find(":") == len(table.iloc[0, 0]) - 1: # check if last character is ":" indicating a vertical table
+                    vertical = True
+                else:
+                    vertical = False
 
         elif cols > 2: # and rows <= 2
             vertical = False
@@ -125,9 +171,12 @@ def parse(filename, output_dir, partnum, dstype):
             vertical = False
         else: # 1 column, <= 2 rows
             vertical = False
-
+        #print(vertical)
         # missing name check
         for table_name_2 in table_list.keys(): 
+            if dstype == "Alphawire" and table_name_2.find("\n") >= 0:
+                torename[table_name_2] = table_name_2[0:table_name_2.find("\n")]
+
             if table_name_2.find(table.iloc[-1, 0]) >= 0:
                 # Name taken from table directly above - this table does not have a name
                 torename[table_name_2] = "Specs " + str(len(tables))
@@ -136,8 +185,12 @@ def parse(filename, output_dir, partnum, dstype):
 
         if vertical:
             out = dict()
+            if rows > 1:
                 for row in table.itertuples(index=False, name=None):
                     out[row[0].replace("\n", " ").replace(":", "")] = row[1]
+            else:
+                for row in table.itertuples(index=False, name=None):
+                    out[row[0].replace("\n", " ").replace(":", "")] = ""
 
         else: # horizontal
             out = dict()
@@ -147,9 +200,52 @@ def parse(filename, output_dir, partnum, dstype):
         
         tables[table_name] = out
 
+        # multi-page table check, Alphawire
+        if dstype == "Alphawire" and table_name.isdigit():
+            # table continues from previous page or has name on previous page
+            thistbl = table_list_raw[table_name]
+            prevtbl = table_list_raw[previous_table]
+            
+            if prevtbl.cells[-1][0].lb[1] < 50 and thistbl.cells[0][0].lt[1] > 600:
+                # wraparound
+                #print("WRAP")
+                #print("PREV TABLE", prevtbl.df)
+                #print("THIS TABLE", thistbl.df)
+                #print("PREV TABLE CORNER", prevtbl.cells[-1][0].lb[1])
+                #print("THIS TABLE CORNER", thistbl.cells[0][0].lt[1])
+                main_key = previous_table
+                cont_key = table_name
+                #print(vertical)
+                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]])
+                            tables[main_key][main_keys[i]] = (tuple(tables[main_key][main_keys[i]]) + (cont_key,) + cont_values)
+    
+                    del tables[table_name]
+    
+                else:
+                    #print(tables[cont_key].keys())
+                    for key in tables[cont_key].keys():
+                        #print(main_key, key, cont_key, key)
+                        tables[main_key][key] = tables[cont_key][key]
+                    del tables[table_name]
+
+            elif thistbl.cells[0][0].lt[1] > 600:
+                # name on previous page (grrrr)
+                #print("NAMEABOVE")
+                #print("PREV TABLE", prevtbl.df)
+                #print("THIS TABLE", thistbl.df)
+                #print("PREV TABLE CORNER", prevtbl.cells[-1][0].lb[1])
+                #print("THIS TABLE CORNER", thistbl.cells[0][0].lt[1])
+                name = extract_table_name(50, prevtbl.page,reader,dstype,table_name).strip("\n").strip()
+                #print("FOUND NAME:", name)
+                torename[table_name] = name
 
 
-        # multi-page table check
+
+        # multi-page table check, Belden
         if dstype == "Belden":
             if table_name.isdigit() and len(tables) > 1:
                 #fprint(table_name)
@@ -177,7 +273,7 @@ def parse(filename, output_dir, partnum, dstype):
 
         previous_table = table_name
     
-    # remove renamed tables
+    # remove & rename tables
     for table_name in torename.keys():
         tables[torename[table_name]] = tables[table_name]
         del tables[table_name]
@@ -211,11 +307,20 @@ def parse(filename, output_dir, partnum, dstype):
 
     #print(output_table)
 
-    run_cmd("rm \"" + output_dir + "\"/*.json") # not reliable!
-    with open(output_dir + "/" + output_table["searchspecs"]["id"] + ".json", 'w') as json_file:
+    #run_cmd("rm \"" + output_dir + "\"/*.json") # not reliable!
+    pattern = os.path.join(output_dir, '*.json')
+    json_files = glob.glob(pattern)
+    for file_path in json_files:
+        os.remove(file_path)
+        #print(f"Deleted {file_path}")
+    with open(output_dir + "/search_" + output_table["searchspecs"]["id"] + ".json", 'w') as json_file:
         json.dump(output_table["searchspecs"], json_file)
-    touch(output_dir + "/parsed")
-    return output_table
+    with open(output_dir + "/specs_" + output_table["partnum"] + ".json", 'w') as json_file:
+        json.dump(output_table["fullspecs"], json_file)
+
+    #print(json.dumps(output_table, indent=2))
+    touch(output_dir + "/parsed") # mark as parsed
+    return True
 
 
 def flatten(tables):
@@ -245,7 +350,6 @@ def flatten(tables):
                 #print("\"" + keyname + "\":", "\"" + str(out[fullkeyname]) + "\",")
             elif len(tables[table][key]) == 1:
                 out[fullkeyname] = convert_to_number(tables[table][key][0])
-                
                 #print("\"" + keyname + "\":", "\"" + str(out[fullkeyname]) + "\",")
 
             # if the item has at least two commas in it, split it
@@ -256,7 +360,7 @@ def flatten(tables):
             # if the item has at least two commas in it, split it
             if tables[table][key].count(',') > 0:
                 out[fullkeyname] = list(map(lambda x: x.strip(), tables[table][key].split(",")))
-                print("\"" + keyname + "\":", "\"" + str(out[fullkeyname]) + "\",")
+                #print("\"" + keyname + "\":", "\"" + str(out[fullkeyname]) + "\",")
 
 
     #print("}")
@@ -265,4 +369,4 @@ def flatten(tables):
     
 
 if __name__ == "__main__":
-    parse("test2.pdf", "cables/10GXS13", "10GXS13")
+    print(parse("cables/3050/datasheet.pdf", "cables/3050", "3050", "Alphawire"))

requirements.txt

@@ -1,5 +1,5 @@
 # Runtime
-camelot-py[base]
+camelot-py
 opencv-python
 pypdf2==2.12.1
 alive-progress
@@ -15,6 +15,13 @@ websockets
 numpy
 scipy
 ipywidgets
+pandas
+pyarrow
+ghostscript
+pyzbar
+
 
 # Development
 matplotlib
+#cx_Freeze # uncomment if building label generator app
+# requires windows 10 SDK, visual C++, etc

setup-label-generator.py

@@ -0,0 +1,31 @@
+import sys
+from cx_Freeze import setup, Executable
+
+debug = True
+debug = not debug
+# Dependencies are automatically detected, but it might need fine tuning.
+# "packages": ["os"] is used as example only
+
+import opcode
+import os
+import distutils
+#distutils_path = os.path.join(os.path.dirname(opcode.__file__), 'distutils')
+build_exe_options = {"include_msvcr": True, "packages": ["camelot", "setuptools"], "optimize": 0, "silent": True, "include_files": ["gs10030w64.exe"], "excludes": ["scipy", "torch"]}
+
+# base="Win32GUI" should be used only for Windows GUI app
+base = "console"
+#if sys.platform == "win32" and not debug:
+#    base = "Win32GUI"
+
+if sys.platform == "linux" or sys.platform == "linux2" or sys.platform == "darwin":
+    name = "jukebox-labelgen"
+else:   
+    name = "jukebox-labelgen.exe"
+
+setup(
+    name="IP Pigeon",
+    version="0.2.4",
+    description="IP Pigeon client application",
+    options={"build_exe": build_exe_options},
+    executables=[Executable("label_generator.py", base=base, uac_admin=False, target_name=name)],
+)

ur5_control.py

@@ -216,35 +216,65 @@ def normalize_degree(theta):
     # Return angle
     return normalized_theta
 
-def get_joints_from_xyz_rel(x, y, z, initial_guess = (math.pi/2, math.pi/2, 0), limbs=(.422864, .359041, .092124)):
-    # Get polar coordinates of x,y pair
-    r, theta = cartesian_to_polar(x, y)
 
-    # Get length of each limb
-    l1, l2, l3 = limbs
+def get_joints_from_xyz_rel(x, y, z, rx=0, ry=-math.pi/2, rz=0, initial_guess = (math.pi/2, math.pi/2, 0)):
+    # Get limbs and offsets
+    offset_x, offset_y, offset_z = (0, 0, 0.14)     # Tool offset
+    l_bs, l1, l2, l3, l_wt = (0.1333, .425, .39225, .1267, .0997)    # Limb lengths
+    #l3=0.15
+    
+    # Calculate base angle and r relative to shoulder joint
+    def calculate_theta(x, y, a):
+        # Calculate if we need the + or - in our equations
+        if (x>-a and y>=0) or (x>a and y<0):
+            flip = 1
+        elif (x<-a and y>=0) or (x<a and y<0):
+            flip = -1
+        else: 
+            # Critical section (x=a, or x=-a). Infinite slope
+            # Return 0 or 180 depending on sign
+            return math.atan2(y, 0)
+        
+        # Calculate tangent line y = mx + b
+        m = (x*y - math.sqrt(x*x*y*y-(x*x-a*a)*(y*y-a*a)))/(x*x-a*a)
+        b = flip * a * math.sqrt(1+m*m)
+
+        # Calculate equivalent tangent point on circle
+        cx = (-flip*m*b)/(1+m*m)
+        cy = m*cx + flip*b
+
+        # Calculate base angle, make angle negative if flip=1
+        theta = math.atan2(cy, cx) + (-math.pi if flip==1 else 0)
+
+        return theta 
+    
+    base_theta = calculate_theta(x, y, l_bs)
+    cx, cy = l_bs*math.cos(base_theta), l_bs*math.sin(l_bs)
+    r = math.sqrt((x-cx)**2 + (y-cy)**2) 
+
 
     # Formulas to find out joint positions for (r, z)
     def inv_kin_r_z(p):
         a, b, c = p            
 
         return (l1*math.cos(a) + l2*math.cos(a-b) + l3*math.cos(a-b-c) - r, # r
-                l1*math.sin(a) + l2*math.sin(a-b) - l3*math.sin(a-b-c) - z,  # z
+                l1*math.sin(a) + l2*math.sin(a-b) - l3*math.sin(a-b-c) - (l3*math.sin(a-b-c)) - (z + offset_z),  # z
                 a-b-c) # wrist angle
 
 
     # Normalize angles
-    base, shoulder, elbow, wrist = [normalize_degree(deg) for deg in [theta, *fsolve(inv_kin_r_z, initial_guess)]]
+    base, shoulder, elbow, wrist1 = [normalize_degree(deg) for deg in [base_theta, *fsolve(inv_kin_r_z, initial_guess)]]
 
     # Return result
-    return base, shoulder, elbow, wrist
+    return base, shoulder, elbow, wrist1, ry, rz
 
-def get_joints_from_xyz_abs(x, y, z):
-    joints = get_joints_from_xyz_rel(x, y, z)
+def get_joints_from_xyz_abs(x, y, z, rx=0, ry=-math.pi/2, rz=math.pi/2):
+    joints = get_joints_from_xyz_rel(x, y, z, rx, ry, rz)
 
     # Joint offsets
     # Base, Shoulder, Elbow, Wrist
-    inverse = [1, -1, 1, 1]
-    offsets = [0, 0, 0, -math.pi/2]
+    inverse = [1, -1, 1, 1, 1, 1]
+    offsets = [-math.pi/2, 0, 0, -math.pi/2, 0, 0]
 
     # Return adjusted joint positions
     return [o+j*i for j, o, i in zip(joints, offsets, inverse)]
@@ -262,7 +292,7 @@ if __name__ == "__main__":
                0.40002172976662786, 
                0, 
                -3.14152741295329, 
-               0]
+               math.radians(62)]
 
     # time.sleep(.5)
 
@@ -290,14 +320,21 @@ if __name__ == "__main__":
     # set_pos_abs(*home_pose)
 
 
-    angles = get_joints_from_xyz_abs(0.3, 0.3, 0.3)
-    rob.movej([*angles, *rob.getj()[4:]], acc=1, vel=1)
     
-    angles = get_joints_from_xyz_abs(-0.3, -0.3, 0.7)
-    rob.movej([*angles, *rob.getj()[4:]], acc=1, vel=1)
+    angles = get_joints_from_xyz_abs(-0.7, 0, 0)
+    rob.movej(angles, acc=2, vel=2)
+
+    # joints = []
+    # for i in np.linspace(-0.3, -0.7, 50):
+    #     joints.append(get_joints_from_xyz_abs(i, 0, 0))
+    # rob.movejs(joints, acc=2, vel=2)
+
+    # time.sleep(5)
+
+
+    # angles = get_joints_from_xyz_abs(0, -0.6, 0)
+    # rob.movej(angles, acc=2, vel=2)
     
-    angles = get_joints_from_xyz_abs(-0.3, 0.4, 0.2)
-    rob.movej([*angles, *rob.getj()[4:]], acc=1, vel=1)
 
 
     # set_pos_abs(*p1)

util.py

@@ -70,7 +70,7 @@ def fprint(msg, settings = None, sendqueue = None):
     except Exception as e:
         try:
             print('[????:' + frm.function + ']:', str(msg))
-            print('[util:fprint]: ' + str(e))
+            #print('[util:fprint]: ' + str(e))
         except:
             print('[????]:', str(msg))