Clean up imports

opentaxforms/Form.py

@@ -1,15 +1,23 @@
+from __future__ import absolute_import
 import re
-try:
-    from urllib2 import urlopen, URLError, HTTPError
-except ImportError:
-    from urllib.request import urlopen
-    from urllib.error import URLError, HTTPError
+from pdfminer.converter import PDFPageAggregator
+from pdfminer.layout import (
+    LTTextBox, LTTextLine, LTTextBoxHorizontal, LAParams, LTChar,
+    LTTextLineHorizontal
+)
+from pdfminer.pdfparser import PDFParser
+from pdfminer.pdfdocument import PDFDocument
+from pdfminer.pdfpage import PDFPage
+from pdfminer.pdftypes import resolve1
+from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
+from six.moves.urllib.request import urlopen
+from six.moves.urllib.error import URLError, HTTPError
 
-import opentaxforms.ut as ut
-from opentaxforms.ut import (log, ntuple, logg, stdout, Qnty, NL,
-    pathjoin)
-import opentaxforms.irs as irs
-from opentaxforms.config import cfg
+from . import ut, irs
+from .ut import log, ntuple, logg, stdout, Qnty, NL, pathjoin
+from .config import cfg
+from .xmp import xmp_to_dict
+from .cmds import CommandParser, normalize, adjustNegativeField, CannotParse
 
 # global so that theyre pickle-able
 PageInfo = ntuple('PageInfo', 'pagenum pagewidth pageheight textpoz')
@@ -133,16 +141,11 @@ def download(self, year, failurls, dirName='forms'):
 
     def pdfInfo(self):
         # collect metadata from pdf file at document and page levels
-        from pdfminer.pdfparser import PDFParser
-        from pdfminer.pdfdocument import PDFDocument
-        from pdfminer.pdfpage import PDFPage
         with open(self.fpath, 'rb') as fp:
             parser = PDFParser(fp)
             doc = PDFDocument(parser)
             docinfo = {}
             if 'Metadata' in doc.catalog:
-                from pdfminer.pdftypes import resolve1
-                from opentaxforms.xmp import xmp_to_dict
                 metadata = resolve1(doc.catalog['Metadata']).get_data()
                 xmpdict = xmp_to_dict(metadata)
                 docinfo['titl'] = xmpdict['dc']['title']['x-default']
@@ -233,8 +236,6 @@ def orderDependencies(self):
     def computeMath(self):
         # determines which fields are computed from others
         # 'dep' means dependency
-        from opentaxforms.cmds import (
-            CommandParser, normalize, adjustNegativeField, CannotParse)
         fields = self.fields if 'm' in cfg.steps else []
         for field in fields:
             math = CommandParser(field, self)
@@ -258,9 +259,6 @@ def computeMath(self):
 
 class Renderer(object):
     def __init__(self):
-        from pdfminer.layout import LAParams
-        from pdfminer.converter import PDFPageAggregator
-        from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
         # Create a PDF resource manager object that stores shared resources.
         rsrcmgr = PDFResourceManager()
         # la=layout analysis
@@ -270,21 +268,19 @@ def __init__(self):
         self.textPoz = None
 
     def renderPage(self, page):
-        from pdfminer.layout import LTTextBox, LTTextLine, LTTextBoxHorizontal
         self.interpreter.process_page(page)
         layout = self.device.get_result()
         # http://denis.papathanasiou.org/2010/08/04/extracting-text-images-
         # from-pdf-files/
         textPoz = TextPoz()
         for lt in layout:
-            if lt.__class__ in (LTTextBoxHorizontal, LTTextBox, LTTextLine):
+            if isinstance(lt, (LTTextBoxHorizontal, LTTextBox, LTTextLine)):
                 textPoz.add(lt)
         return textPoz
 
 
 class TextPoz(object):
     # text positions
-    from pdfminer.layout import LTChar, LTTextLineHorizontal
     FormPos = ntuple('FormPos', 'itxt ichar chrz bbox')
 
     def __init__(self):
@@ -296,13 +292,13 @@ def quantify(tupl, unit):
 
         def accum(ltobj, ltchars, chars):
             for lto in ltobj:
-                if isinstance(lto, self.LTChar):
+                if isinstance(lto, LTChar):
                     ltchartext = lto.get_text()
                     ltchars.append(
                         (ltchartext,
                          ut.Bbox(*quantify(lto.bbox, 'printers_point'))))
                     chars.append(ltchartext)
-                elif isinstance(lto, self.LTTextLineHorizontal):
+                elif isinstance(lto, LTTextLineHorizontal):
                     accum(lto, ltchars, chars)
         ltchars = []
         chars = []

opentaxforms/cmds.py

@@ -1,8 +1,9 @@
+from __future__ import absolute_import
 import re
 import six
-import opentaxforms.ut as ut
-from opentaxforms.ut import log, jj, numerify
-import opentaxforms.irs as irs
+
+from . import ut, irs
+from .ut import log, jj, numerify
 
 
 def normalize(s):

opentaxforms/config.py

@@ -1,11 +1,17 @@
-from __future__ import print_function
-import sys
+from __future__ import print_function, absolute_import
 import os
 import os.path
+import re
+import shutil
+import sys
 from argparse import ArgumentParser
-import opentaxforms.ut as ut
-from opentaxforms.ut import log, Bag, setupLogging, logg, NL, pathjoin
-from opentaxforms.version import appname, appversion
+from os.path import isfile, join as joinpath
+from six.moves.urllib.request import urlopen
+from six.moves.urllib.error import URLError
+
+from . import ut
+from .ut import log, Bag, setupLogging, logg, NL, pathjoin
+from .version import appname, appversion
 
 RecurseInfinitely = -1
 RecursionRootLevel = 0
@@ -123,15 +129,13 @@ def getFileList(dirName):
             except Exception as e:
                 log.warn('cannot symlink %s to %s because %s, copying instead'%(
                     allpdfpath, allpdfLink, e, ))
-                import shutil
                 shutil.copy(allpdfpath,allpdfLink)
         elif not cfg.okToDownload:
             msg = 'allPdfNames file [%s] not found but dontDownload' % (
                 allpdfpath)
             raise Exception(msg)
         else:
             # todo why did this stop working?  my own env?
-            from urllib2 import urlopen, URLError
             try:
                 # could use https://www.irs.gov/pub/irs-pdf/pdfnames.txt but
                 # this way we avoid errors in that file
@@ -141,7 +145,6 @@ def getFileList(dirName):
                         fin.getcode(), ))
                 allpdffiles_html = fin.read()
                 fin.close()
-                import re
                 allpdfnames = re.findall(r'f[\w\d-]+\.pdf', allpdffiles_html)
                 allpdfnames = ut.uniqify(sorted(allpdfnames))
                 with open(allpdfpath, 'w') as f:
@@ -212,12 +215,12 @@ def setup(**overrideArgs):
         log.warn('commandline: %s at %s', ' '.join(sys.argv), ut.now())
 
     if dirName is not None:
-        from opentaxforms.Form import Form
+        # deferred import to avoid circular reference
+        from .Form import Form
         if rootForms:
             cfg.formsRequested = [
                 Form(rootForm, RecursionRootLevel) for rootForm in rootForms]
         else:
-            from os.path import isfile, join as joinpath
             cfg.formsRequested = [
                 Form(f, RecursionRootLevel)
                 for f in os.listdir(dirName)
@@ -235,15 +238,13 @@ def setup(**overrideArgs):
         ut.ensure_dir(dirName)
         staticDir = ut.Resource(appname, 'static').path()
         staticLink = pathjoin(dirName, 'static')
-        import os.path
         try:
             if not os.path.lexists(staticLink):
                 os.symlink(staticDir, staticLink)
         except Exception as e:
             log.warn('cannot symlink %s to %s because %s, copying instead'%(
                 staticDir, staticLink, e))
             try:
-                import shutil
                 shutil.copytree(staticDir, staticLink)
             except Exception as e:
                 log.warn('cannot copy %s to %s because %s,'

opentaxforms/db.py

@@ -1,13 +1,16 @@
+from __future__ import absolute_import
+import os
 import six
+import sys
 from itertools import chain
-import opentaxforms.ut as ut
-import opentaxforms.config as config
-from opentaxforms.ut import log
-from opentaxforms.config import cfg
-from sqlalchemy import MetaData, create_engine, select
-from sqlalchemy import UniqueConstraint
+from sqlalchemy import (
+    MetaData, create_engine, select, UniqueConstraint)
 # from sqlalchemy.exc import ProgrammingError
 
+from . import ut, config
+from .ut import log
+from .config import cfg
+
 engine, metadata, conn = None, None, None
 
 
@@ -23,7 +26,6 @@ def connect(appname, **kw):
     user = pw = 'user'
     dbname = appname.lower()
     # optionally override defaults
-    import os
     user = os.environ.get(appname.upper() + '_DBUSER', user)
     pw = os.environ.get(appname.upper() + '_DBPASS', pw)
     dbname = os.environ.get(appname.upper() + '_DBNAME', dbname)
@@ -240,7 +242,6 @@ def stripifstring(s):
 
 
 if __name__ == "__main__":
-    import sys
     args = sys.argv
     if any([arg in args for arg in '-t --testing'.split()]):
         import doctest

opentaxforms/html.py

@@ -1,11 +1,14 @@
-from __future__ import print_function
-from os import remove as removeFile
+from __future__ import print_function, absolute_import
+import os.path
 import re
+import traceback
+from os import remove as removeFile
 from itertools import chain
-from opentaxforms.config import cfg,setup
-from opentaxforms.irs import computeTitle, computeFormId, sortableFieldname
-import opentaxforms.ut as ut
-from opentaxforms.ut import log, jdb, Qnty, NL, pathjoin
+
+from . import ut
+from .config import cfg, setup
+from .irs import computeTitle, computeFormId, sortableFieldname
+from .ut import log, jdb, Qnty, NL, pathjoin
 
 
 def computeFormFilename(form):
@@ -44,7 +47,6 @@ def computePageTitle(titlebase, npage, npages):
 
 def createSvgFile(dirName, prefix, npage):
     ipage = npage - 1
-    import os.path
     infpath = os.path.join(dirName,'{}.pdf'.format(prefix))
     print('infpath', infpath)
     outfpath = os.path.join(dirName,'{}-p{}-fixedDims.svg'.format(prefix, ipage))
@@ -75,6 +77,15 @@ def createSvgFile(dirName, prefix, npage):
     removeFile(outfpath)
 
 
+def readImgSize(fname, dirName):
+    # deferred import: PIL/Pillow isn't a hard dependency
+    from PIL import Image
+    with open(pathjoin(dirName,fname), 'rb') as fh:
+        img = Image.open(fh)
+    # (width, height) in pixels
+    return img.size
+
+
 def createGifFile(dirName, prefix, npage):
     ipage = npage - 1
     imgfname = prefix + '.gif'
@@ -92,7 +103,7 @@ def createGifFile(dirName, prefix, npage):
     try:
         imgw, imgh = ut.readImgSize(imgfname, dirName)
     except:
-        log.error('err re file ' + imgfname + ' in dir ' + dirName)
+        log.error('err re file %s in dir %s', imgfname, dirName)
         raise
     return imgw, imgh
 
@@ -260,7 +271,6 @@ def titleValue(f):
                     shorten(f.xpos),
                     shorten(f.ypos)) if tooltip else '')
         except Exception:
-            import traceback
             log.warn(ut.jj('caughtError:', traceback.format_exc()))
             return ''
 

opentaxforms/link.py

@@ -1,7 +1,9 @@
+from __future__ import absolute_import
 import six
 import re
-from opentaxforms.ut import log, jj, ddict
-import opentaxforms.irs as irs
+
+from . import irs
+from .ut import log, jj, ddict
 
 
 def findLineAndUnit(speak):

opentaxforms/main.py

@@ -10,14 +10,16 @@
     pos,poz=position,positions
 '''
 from __future__ import print_function, absolute_import
+import json
 import sys
-from os import remove as removeFile
 import traceback
+from os import remove as removeFile
 
 from . import ut, irs, link, schema, html, refs as references
 from .ut import log, jj, Bag, logg, stdout, Qnty, pathjoin
 from .config import cfg, setup, RecurseInfinitely
 from .extractFillableFields import extractFields
+from .Form import Form
 
 
 def cleanup_files(form):
@@ -33,7 +35,6 @@ def cleanup_files(form):
 
 
 def addFormsTodo(form, formsdone, formstodo, formsfail):
-    from opentaxforms.Form import Form
     recurselevel = form.recurselevel
     refs = form.refs
     if cfg.recurse and (cfg.maxrecurselevel == RecurseInfinitely or
@@ -129,7 +130,6 @@ def logRunStatus(formsdone, formsfail, status):
         msg = 'failed to process %d forms: %s' % (
               len(formsfail), [irs.computeFormId(f) for f in formsfail])
         logg(msg, [log.error, stdout])
-    import json
     status.update({'f' + irs.computeFormId(f).lower(): None
                   for f in formsfail})
     statusStr = json.dumps(status.__dict__)

opentaxforms/refs.py

@@ -1,9 +1,10 @@
+from __future__ import absolute_import
 import re
 import six
-import opentaxforms.ut as ut
-from opentaxforms.ut import log, jj, pathjoin
-from opentaxforms.config import cfg
-import opentaxforms.irs as irs
+
+from . import ut, irs
+from .ut import log, jj, pathjoin
+from .config import cfg
 
 # nonforms are numbers that dont represent forms
 nonforms = [str(yr) for yr in range(2000, 2050)]