Fix CLI

camelot/cli.py

@@ -5,10 +5,20 @@ import click
 
 from . import __version__
 from .io import read_pdf
-from .utils import validate_input, remove_extra
 
 
-@click.command()
+class Config(object):
+    def  __init__(self):
+        self.config = {}
+
+    def set_config(self, key, value):
+        self.config[key] = value
+
+
+pass_config = click.make_pass_decorator(Config)
+
+
+@click.group()
 @click.version_option(version=__version__)
 @click.option("-p", "--pages", default="1", help="Comma-separated page numbers"
               " to parse. Example: 1,3,4 or 1,4-end")
@@ -18,9 +28,6 @@ from .utils import validate_input, remove_extra
               help="Output file format.")
 @click.option("-z", "--zip", is_flag=True, help="Whether or not to create a ZIP"
               " archive.")
-@click.option("-T", "--table_area", default=[], multiple=True,
-              help="Table areas (x1,y1,x2,y2) to process.\n"
-              " x1, y1 -> left-top and x2, y2 -> right-bottom")
 @click.option("-split", "--split_text", is_flag=True, help="Whether or not to"
               " split text if it spans across multiple cells.")
 @click.option("-flag", "--flag_size", is_flag=True, help="(inactive) Whether or"
@@ -28,86 +35,121 @@ from .utils import validate_input, remove_extra
               " super/subscripts)")
 @click.option("-M", "--margins", nargs=3, default=(1.0, 0.5, 0.1),
               help="char_margin, line_margin, word_margin for PDFMiner.")
-@click.option("-G", "--geometry_type",
+@click.pass_context
+def cli(ctx, *args, **kwargs):
+    ctx.obj = Config()
+    for key, value in kwargs.iteritems():
+        ctx.obj.set_config(key, value)
+
+
+@cli.command('lattice')
+@click.option("-T", "--table_area", default=[], multiple=True,
+              help="Table areas (x1,y1,x2,y2) to process.\n"
+              " x1, y1 -> left-top and x2, y2 -> right-bottom")
+@click.option("-back", "--process_background", is_flag=True,
+              help="(with --mesh) Whether or not to process lines that are in"
+              " background.")
+@click.option("-scale", "--line_size_scaling", default=15,
+              help="(with --mesh) Factor by which the page dimensions will be"
+              " divided to get smallest length of detected lines.")
+@click.option("-copy", "--copy_text", default=[], type=click.Choice(["h", "v"]),
+              multiple=True, help="(with --mesh) Specify direction"
+              " in which text will be copied over in a spanning cell.")
+@click.option("-shift", "--shift_text", default=["l", "t"],
+              type=click.Choice(["", "l", "r", "t", "b"]), multiple=True,
+              help="(with --mesh) Specify direction in which text in a spanning"
+              " cell should flow.")
+@click.option("-l", "--line_close_tol", default=2,
+              help="(with --mesh) Tolerance parameter used to merge close vertical"
+              " lines and close horizontal lines.")
+@click.option("-j", "--joint_close_tol", default=2,
+              help="(with --mesh) Tolerance parameter used to decide whether"
+              " the detected lines and points lie close to each other.")
+@click.option("-block", "--threshold_blocksize", default=15,
+              help="(with --mesh) For adaptive thresholding, size of a pixel"
+              " neighborhood that is used to calculate a threshold value for"
+              " the pixel: 3, 5, 7, and so on.")
+@click.option("-const", "--threshold_constant", default=-2,
+              help="(with --mesh) For adaptive thresholding, constant subtracted"
+              " from the mean or weighted mean.\nNormally, it is positive but"
+              " may be zero or negative as well.")
+@click.option("-I", "--iterations", default=0,
+              help="(with --mesh) Number of times for erosion/dilation is"
+              " applied.")
+@click.option("-plot", "--plot_type",
               type=click.Choice(["text", "table", "contour", "joint", "line"]),
-              help="Plot geometry found on pdf page for debugging.\n\n"
+              help="Plot geometry found on PDF page for debugging.")
-              "text: Plot text objects. (Useful to get table_area and"
-              " columns coordinates)\ntable: Plot parsed table.\n"
-              "contour (with --mesh): Plot detected rectangles.\njoint (with --mesh): Plot detected line"
-              " intersections.\nline (with --mesh): Plot detected lines.")
 @click.argument("filepath", type=click.Path(exists=True))
-def cli(*args, **kwargs):
+@pass_config
-    pages = kwargs.pop("pages")
+def lattice(c, *args, **kwargs):
-    output = kwargs.pop("output")
+    """Use lines between text to generate table."""
-    f = kwargs.pop("format")
+    conf = c.config
-    compress = kwargs.pop("zip")
+    pages = conf.pop("pages")
-    mesh = kwargs.pop("mesh")
+    output = conf.pop("output")
-    geometry_type = kwargs.pop("geometry_type")
+    f = conf.pop("format")
+    compress = conf.pop("zip")
+    plot_type = kwargs.pop('plot_type')
     filepath = kwargs.pop("filepath")
+    kwargs.update(conf)
+
+    table_area = list(kwargs['table_area'])
+    kwargs['table_area'] = None if not table_area else table_area
+    copy_text = list(kwargs['copy_text'])
+    kwargs['copy_text'] = None if not copy_text else copy_text
+    kwargs['shift_text'] = list(kwargs['shift_text'])
+
+    tables = read_pdf(filepath, pages=pages, flavor='lattice', **kwargs)
+    click.echo(tables)
+    if plot_type is not None:
+        for table in tables:
+            table.plot(plot_type)
+    else:
+        if output is None:
+            raise click.UsageError("Please specify output filepath using --output")
+        if f is None:
+            raise click.UsageError("Please specify output format using --format")
+        tables.export(output, f=f, compress=compress)
+
+
+@cli.command('stream')
+@click.option("-T", "--table_area", default=[], multiple=True,
+              help="Table areas (x1,y1,x2,y2) to process.\n"
+              " x1, y1 -> left-top and x2, y2 -> right-bottom")
+@click.option("-C", "--columns", default=[], multiple=True,
+              help="x-coordinates of column separators.")
+@click.option("-r", "--row_close_tol", default=2, help="Rows will be"
+              " formed by combining text vertically within this tolerance.")
+@click.option("-c", "--col_close_tol", default=0, help="Columns will"
+              " be formed by combining text horizontally within this tolerance.")
+@click.option("-plot", "--plot_type",
+              type=click.Choice(["text", "table"]),
+              help="Plot geometry found on PDF page for debugging.")
+@click.argument("filepath", type=click.Path(exists=True))
+@pass_config
+def stream(c, *args, **kwargs):
+    """Use spaces between text to generate table."""
+    conf = c.config
+    pages = conf.pop("pages")
+    output = conf.pop("output")
+    f = conf.pop("format")
+    compress = conf.pop("zip")
+    plot_type = kwargs.pop('plot_type')
+    filepath = kwargs.pop("filepath")
+    kwargs.update(conf)
 
     table_area = list(kwargs['table_area'])
     kwargs['table_area'] = None if not table_area else table_area
     columns = list(kwargs['columns'])
     kwargs['columns'] = None if not columns else columns
-    copy_text = list(kwargs['copy_text'])
-    kwargs['copy_text'] = None if not copy_text else copy_text
-    kwargs['shift_text'] = list(kwargs['shift_text'])
 
-    kwargs = remove_extra(kwargs, mesh=mesh)
+    tables = read_pdf(filepath, pages=pages, flavor='stream', **kwargs)
-    tables = read_pdf(filepath, pages=pages, mesh=mesh, **kwargs)
     click.echo(tables)
-    if output is None:
+    if plot_type is not None:
-        raise click.UsageError("Please specify an output filepath using --output")
+        for table in tables:
-    if f is None:
+            table.plot(plot_type)
-        raise click.UsageError("Please specify an output format using --format")
+    else:
-    tables.export(output, f=f, compress=compress)
+        if output is None:
-
+            raise click.UsageError("Please specify output filepath using --output")
-
+        if f is None:
-@click.option("-T", "--table_area", default=[], multiple=True,
+            raise click.UsageError("Please specify output format using --format")
-              help="Table areas (x1,y1,x2,y2) to process.\n"
+        tables.export(output, f=f, compress=compress)
-              " x1, y1 -> left-top and x2, y2 -> right-bottom")
-@click.option("-back", "--process_background", is_flag=True, cls=Mutex,
-              help="(with --mesh) Whether or not to process lines that are in"
-              " background.")
-@click.option("-scale", "--line_size_scaling", default=15, cls=Mutex,
-              help="(with --mesh) Factor by which the page dimensions will be"
-              " divided to get smallest length of detected lines.")
-@click.option("-copy", "--copy_text", default=[], type=click.Choice(["h", "v"]),
-              multiple=True, cls=Mutex, help="(with --mesh) Specify direction"
-              " in which text will be copied over in a spanning cell.")
-@click.option("-shift", "--shift_text", default=["l", "t"],
-              type=click.Choice(["", "l", "r", "t", "b"]), multiple=True, cls=Mutex,
-              help="(with --mesh) Specify direction in which text in a spanning"
-              " cell should flow.")
-@click.option("-l", "--line_close_tol", default=2, cls=Mutex,
-              help="(with --mesh) Tolerance parameter used to merge close vertical"
-              " lines and close horizontal lines.")
-@click.option("-j", "--joint_close_tol", default=2, cls=Mutex,
-              help="(with --mesh) Tolerance parameter used to decide whether"
-              " the detected lines and points lie close to each other.")
-@click.option("-block", "--threshold_blocksize", default=15, cls=Mutex,
-              help="(with --mesh) For adaptive thresholding, size of a pixel"
-              " neighborhood that is used to calculate a threshold value for"
-              " the pixel: 3, 5, 7, and so on.")
-@click.option("-const", "--threshold_constant", default=-2, cls=Mutex,
-              help="(with --mesh) For adaptive thresholding, constant subtracted"
-              " from the mean or weighted mean.\nNormally, it is positive but"
-              " may be zero or negative as well.")
-@click.option("-I", "--iterations", default=0, cls=Mutex,
-              help="(with --mesh) Number of times for erosion/dilation is"
-              " applied.")
-def lattice(*args, **kwargs):
-    pass
-
-
-@click.option("-T", "--table_area", default=[], multiple=True,
-              help="Table areas (x1,y1,x2,y2) to process.\n"
-              " x1, y1 -> left-top and x2, y2 -> right-bottom")
-@click.option("-C", "--columns", default=[], multiple=True, cls=Mutex,
-              help="x-coordinates of column separators.")
-@click.option("-r", "--row_close_tol", default=2, cls=Mutex, help="Rows will be"
-              " formed by combining text vertically within this tolerance.")
-@click.option("-c", "--col_close_tol", default=0, cls=Mutex, help="Columns will"
-              " be formed by combining text horizontally within this tolerance.")
-def stream(*args, **kwargs):
-    pass

camelot/parsers/lattice.py

@@ -21,7 +21,7 @@ logger = setup_logging(__name__)
 
 class Lattice(BaseParser):
     """Lattice method of parsing looks for lines between text
-    to form a table.
+    to generate table.
 
     Parameters
     ----------
@@ -83,7 +83,7 @@ class Lattice(BaseParser):
                  line_size_scaling=15, copy_text=None, shift_text=['l', 't'],
                  split_text=False, flag_size=False, line_close_tol=2,
                  joint_close_tol=2, threshold_blocksize=15, threshold_constant=-2,
-                 iterations=0, margins=(1.0, 0.5, 0.1)):
+                 iterations=0, margins=(1.0, 0.5, 0.1), **kwargs):
         self.table_area = table_area
         self.process_background = process_background
         self.line_size_scaling = line_size_scaling

camelot/parsers/stream.py

@@ -16,7 +16,7 @@ logger = setup_logging(__name__)
 
 class Stream(BaseParser):
     """Stream method of parsing looks for spaces between text
-    to form a table.
+    to generate table.
 
     If you want to specify columns when specifying multiple table
     areas, make sure that the length of both lists are equal.
@@ -51,7 +51,7 @@ class Stream(BaseParser):
     """
     def __init__(self, table_area=None, columns=None, split_text=False,
                  flag_size=False, row_close_tol=2, col_close_tol=0,
-                 margins=(1.0, 0.5, 0.1)):
+                 margins=(1.0, 0.5, 0.1), **kwargs):
         self.table_area = table_area
         self.columns = columns
         self._validate_columns()