3

t2�e1R�@szdZd
Z
ddlZddlZddlmZ
ddlmZ
Gdd�de�ZGdd	�d	�Z	Gd
d�de	�Z
Gdd
�d
e	�Zdd�ZdS)a�

This module defines table parser classes,which parse plaintext-graphic tables
and produce a well-formed data structure suitable for building a CALS table.

:Classes:
    - `GridTableParser`: Parse fully-formed tables represented with a grid.
    - `SimpleTableParser`: Parse simple tables, delimited by top & bottom
      borders.

:Exception class: `TableMarkupError`

:Function:
    `update_dict_of_lists()`: Merge two dictionaries containing list values.
ZreStructuredText�N)
�	DataError)
�strip_combining_charsc@seZ
dZd
Zdd�ZdS)�TableMarkupErrorz�
    Raise if there is any problem with table markup.

    The keyword argument `offset` denotes the offset of the problem
    from the table's start line.
    c
Os"|jd
d�|_
tj|f
|
��
dS)N�offsetr
)�poprr�__init__)�self�args�kwargs�r�!/usr/lib/python3.6/tableparser.pyr&s

zTableMarkupError.__init__N)�__name__�
__module__�__qualname__�__doc__rrrrrrsrc@s(eZ
dZd
ZdZdZdd�Zdd�ZdS)�TableParserzR
    Abstract superclass for the common parts of the syntax-specific parsers.
    N�
cCs&|j|
�

|j
�
|j�
|j�}|S)
an

        Analyze the text `block` and return a table data structure.

        Given a plaintext-graphic table in `block` (list of lines of text; no
        whitespace padding), parse the table, construct and return the data
        necessary to construct a CALS table or equivalent.

        Raise `TableMarkupError` if there is any problem with the markup.
        )�setup�find_head_body_sep�parse_table�structure_from_cells)r�blockZ	structurerrr�parse7s






zTableParser.parsec
Cs�xjtt
|j�
�
D]X}
|j|
}|jj|�
r|jrPtd
|jd|
df|
d��
q|
|_|jdd�|j|
<qW|jdks�|jt
|j�
dkr�td|
d��
dS)	z>Look for a head/body row separator line; store the line index.zLMultiple head/body row separators (table lines %s and %s); only one allowed.�
)
r�
=�
-r
zKThe head/body row separator may not be the first or last line of the table.N)�range�lenr�head_body_separator_pat�match�
head_body_sepr�replace)r�
i�linerrrrGs









zTableParser.find_head_body_sep)r
rrrr�double_width_pad_charrrrrrrr+s
rc@sjeZ
dZd
Zejd�
Zdd�Zdd�Zdd�Z	d	d
�Z
dd�Zd
d�Zdd�Z
dd�Zdd�Zdd�ZdS)�GridTableParserai	
    Parse a grid table using `parse()`.

    Here's an example of a grid table::

        +------------------------+------------+----------+----------+
        | Header row, column 1   | Header 2   | Header 3 | Header 4 |
        +========================+============+==========+==========+
        | body row 1, column 1   | column 2   | column 3 | column 4 |
        +------------------------+------------+----------+----------+
        | body row 2             | Cells may span columns.          |
        +------------------------+------------+---------------------+
        | body row 3             | Cells may  | - Table cells       |
        +------------------------+ span rows. | - contain           |
        | body row 4             |            | - body elements.    |
        +------------------------+------------+---------------------+

    Intersections use '+', row separators use '-' (except for one optional
    head/body row separator, which uses '='), and column separators use '|'.

    Passing the above table to the `parse()` method will result in the
    following data structure::

        ([24, 12, 10, 10],
         [[(0, 0, 1, ['Header row, column 1']),
           (0, 0, 1, ['Header 2']),
           (0, 0, 1, ['Header 3']),
           (0, 0, 1, ['Header 4'])]],
         [[(0, 0, 3, ['body row 1, column 1']),
           (0, 0, 3, ['column 2']),
           (0, 0, 3, ['column 3']),
           (0, 0, 3, ['column 4'])],
          [(0, 0, 5, ['body row 2']),
           (0, 2, 5, ['Cells may span columns.']),
           None,
           None],
          [(0, 0, 7, ['body row 3']),
           (1, 0, 7, ['Cells may', 'span rows.', '']),
           (1, 1, 7, ['- Table cells', '- contain', '- body elements.']),
           None],
          [(0, 0, 9, ['body row 4']), None, None, None]])

    The first item is a list containing column widths (colspecs). The second
    item is a list of head rows, and the third is a list of body rows. Each
    row contains a list of cells. Each cell is either None (for a cell unused
    because of another cell's span), or a tuple. A cell tuple contains four
    items: the number of extra rows used by the cell in a vertical span
    (morerows); the number of extra columns used by the cell in a horizontal
    span (morecols); the line offset of the first line of the cell contents;
    and the cell contents, a list of lines of text.
    z\+=[=+]+=\+ *$cCst|
dd�|_|jj
�
t|
�
d
|_t|
d�
d
|_d|_dg
t|
d�
|_g|_ddg
i
|_ddg
i
|_	dS)Nrr
���)
r�
disconnectr�bottom�rightr �done�cells�rowseps�colseps)rrrrrr�s









zGridTableParser.setupc

Cs
dg
}
x�|
r�|
jd
�
\}}||j
ks||jks||j|k
r>q|j||�}|sPq|\}}}}t|j|�
t|j|�
|j||||�
|j	j
|d|d||�}	|	j�
|	j|j
d�
|jj|||||	f�

|
j||f||fg�

|
j�
qW|j��s�td�
�
dS)a	
        Start with a queue of upper-left corners, containing the upper-left
        corner of the table itself. Trace out one rectangular cell, remember
        it, and add its upper-right and lower-left corners to the queue of
        potential upper-left corners of further cells. Process the queue in
        top-to-bottom order, keeping track of how much of each text column has
        been seen.

        We'll end up knowing all the row and column boundaries, cell positions
        and their dimensions.
        r
r�z"Malformed table; parse incomplete.N)r
r
)rr(r)r*�	scan_cell�update_dict_of_listsr,r-�	mark_doner�get_2D_blockr'r!r$r+�append�extend�sort�check_parse_completer)
rZcorners�top�left�resultr(r)r,r-�	cellblockrrrr�s,





















zGridTableParser.parse_tablecCs2|
d
}|d
}xt||�D]}||j
|<qWdS)z@For keeping track of how much of each text column has been seen.rN)rr*)rr7r8r(r)Zbefore�after�colrrrr1�s

zGridTableParser.mark_donec
Cs4|jd
}
x$t
|j�
D]}|j||
krdSqWdS)z2Each text column should have been completely seen.rFT)r(rr)r*)rZlastr<rrrr6�s





z$GridTableParser.check_parse_completecCs|j|
|�}|S)
z:Starting at the top-left corner, start tracing out a cell.)
�
scan_right)rr7r8r9rrrr/�s
zGridTableParser.scan_cellc
Cs�i}|j|
}xvt
|d
|jd
�D]^}||dkrr|
g
||<|j|
||�}|r�|\}}}	t||	�
||||fSq$||dkr$dSq$WdS)zr
        Look for the top-right corner of the cell, and make note of all column
        boundaries ('+').
        r�
+rN)rrr)�	scan_downr0)
rr7r8r-r#r"r9r(r,Z
newcolsepsrrrr=�s















zGridTableParser.scan_rightc	Cs�i}x�t|
d
|j
d
�D]h}|j||dkrl|g
||<|j|
|||�}|r�|\}}t||�
|||fSq|j||dkrdSqWdS)zj
        Look for the bottom-right corner of the cell, making note of all row
        boundaries.
        rr>�
|N)rr(r�	scan_leftr0)	rr7r8r)r,r"r9Z
newrowsepsr-rrrr?�s












zGridTableParser.scan_downc
Cs�i}|j|}x@t
|d
|d�D],}||dkr<|g
||<q ||dkr dSq W||dkr`dS|j|
|||�}|dk	r�|}	|	|fSdS)z�
        Noting column boundaries, look for the bottom-left corner of the cell.
        It must line up with the starting point.
        rr>rNr&)rr�scan_up)
rr7r8r(r)r-r#r"r9r,rrrrA�s













zGridTableParser.scan_leftcCsVi}xLt|d
|
d�D]8}|j
||dkr8|g
||<q|j
||dkrdSqW|S)zT
        Noting row boundaries, see if we can return to the starting point.
        rr>r@Nr&)rr)rr7r8r(r)r,r"rrrrB
s





zGridTableParser.scan_upc
s�
t|j
j��
}
|
j�
i}x tt|
�
�
D]}|||
|<q(Wt|jj��
��j�
i}x tt��
�
D]}||�|<qdW�f
d
d�tdt��
�D�
}dd�tt��
d�
D�
�
�
f
dd�tt|
�
d�
D�
}t|
�
dt��
d}xt|jD]j\}}	}
}}||}
||	}||
|
d}|||d}||d|d8}|||d|f||
|<q�W|j�
r�||j}|d|�}||d�}ng}|}|||fS)zh
        From the data collected by `scan_cell()`, convert to the final data
        structure.
        c
s$g|]}
�|
�|
dd�qS)
rr)�.0r")
r-rr�
<listcomp>+
s
z8GridTableParser.structure_from_cells.<locals>.<listcomp>rc
Ssg|]}
d�qS)
Nr)rCr"rrrrD.
sc
sg|]}
�dd��qS)
Nr)rCr")
�onerowrrrD/
sN)	�listr,�keysr5rrr-r+r )rr,Zrowindexr"Zcolindex�colspecsZrowsZ	remainingr7r8r(r)rZrownumZcolnumZmorerows�morecolsZnumheadrowsZheadrowsZbodyrowsr)r-rErr
s:























z$GridTableParser.structure_from_cellsN)r
rrr�re�compilerrrr1r6r/r=r?rArBrrrrrr%[s4
#r%c@s^eZ
dZd
Zejd�
Zejd�
Zdd�Zdd�Z	dd	�Z
d
d�Zdd
d�
Zdd�Z
dd�ZdS)�SimpleTableParseraQ
    Parse a simple table using `parse()`.

    Here's an example of a simple table::

        =====  =====
        col 1  col 2
        =====  =====
        1      Second column of row 1.
        2      Second column of row 2.
               Second line of paragraph.
        3      - Second column of row 3.

               - Second item in bullet
                 list (row 3, column 2).
        4 is a span
        ------------
        5
        =====  =====

    Top and bottom borders use '=', column span underlines use '-', column
    separation is indicated with spaces.

    Passing the above table to the `parse()` method will result in the
    following data structure, whose interpretation is the same as for
    `GridTableParser`::

        ([5, 25],
         [[(0, 0, 1, ['col 1']),
           (0, 0, 1, ['col 2'])]],
         [[(0, 0, 3, ['1']),
           (0, 0, 3, ['Second column of row 1.'])],
          [(0, 0, 4, ['2']),
           (0, 0, 4, ['Second column of row 2.',
                      'Second line of paragraph.'])],
          [(0, 0, 6, ['3']),
           (0, 0, 6, ['- Second column of row 3.',
                      '',
                      '- Second item in bullet',
                      '  list (row 3, column 2).'])],
          [(0, 1, 10, ['4 is a span'])],
          [(0, 0, 12, ['5']),
           (0, 0, 12, [''])]])
    z=[ =]*$z-[ -]*$cCs�|
dd�|_|jj
�
|jd
jdd�|jd
<|jdjdd�|jd<d|_g|_d|_g|_dg
t|
d
�
|_d
d
g
i
|_	d
d
g
i
|_
dS)Nr
rrrr&r&r&)rr'r!r �columns�
border_end�tablerr*r,r-)rrrrrrz
s










zSimpleTableParser.setupc
Cs�|j|j
d
d
�|_|jdd|_|jd
\}
}d}d}d}x�|t|j
�
kr�|j
|}|jj|�
r�|j|j
||�||j�|f�
|d}d}nH||
|�j	�r�|r�||kr�|j|j
||�|�
|}d}n|s�|d}|d7}q@WdS)z�
        First determine the column boundaries from the top border, then
        process rows.  Each row may consist of multiple lines; accumulate
        lines until a row is complete.  Call `self.parse_row` to finish the
        job.
        r
rNr&)
�
parse_columnsrrMrNr�span_patr�	parse_row�rstrip�strip)rZ
firststartZfirstendr�startZ
text_foundr#rrrr�
s*


















zSimpleTableParser.parse_tablecCs�g}d
}xD|
jd|�}|
jd|�}|d
kr,P|d
kr<t
|
�
}|j||f�

q
W|jr�|dd|jkrztd|d|d��
|dd
|jd	df|d
<|S)zU
        Given a column span underline, return a list of (begin, end) pairs.
        r
r�
 rz(Column span incomplete in table line %s.)
rr&r&r&r&)�findrr3rMrNr)rr#rZcols�endZbeginrrrrP�
s"














zSimpleTableParser.parse_columnscCs�d
}g}x�|
D]�\}}d
}y,x&||j|dkrB|d7}|d7}qWWn0t
tfk
rv


td|d|dd��
YnX|jd
||gg�

|d7}qW|S)Nr
rz/Column span alignment problem in table line %s.�)
r)rM�AssertionError�
IndexErrorrr3)rZcolspecrr"r+rUrXrIrrr�init_row�
s













zSimpleTableParser.init_rowNc
Cs�|
p|sd
S|r$|j|�}|d}n|j
d
d
�}|}|j|
||�
|j||�}xVtt|�
�
D]F}||\}}|
jd|t|
�
|�}	|	j�
|	j|j	d�
|	||d<q^W|j
j|�

d
S)af

        Given the text `lines` of a row, parse it and append to `self.table`.

        The row is parsed according to the current column spec (either
        `spanline` if provided or `self.columns`).  For each column, extract
        text from each line, and check for text in column margins.  Finally,
        adjust for insignificant whitespace.
        Nrr
r.�)rPrM�
check_columnsr\rrr2r'r!r$rOr3)
r�linesrUZspanlinerMZspan_offset�rowr"rXr:rrrrR�
s 	














zSimpleTableParser.parse_rowcCs 
|jt
jd
f�

t|�
d}dd�|
D�
}
x�tt|�
d�
D]�}||\}}||dd}d}	x�|
D]�}
||kr�|
|d
�j�r�|
|d
�j�}|t|�
}|jd	\}
}|t||�f||<||kr�|
|f|jd
<n.|
||�j��
rt	d||	d||	d��
|	d7}	qfWq<W|j
�
d
S)z�
        Check for text in column margins and text overflow in the last column.
        Raise TableMarkupError if anything but whitespace is in column margins.
        Adjust the end value for the last column if there is text overflow.
        NrYc
Ssg|]}
t|
�
�qSr)
r)rCr#rrrrD�
sz3SimpleTableParser.check_columns.<locals>.<listcomp>rr
z'Text in column margin in table line %s.)
rr&r&)r3�sys�maxsizerrrTrSrM�maxrr)rr_Z
first_linerMZlastcolr"rUrXZ	nextstartrr#�textZnew_endZ
main_startZmain_endrrrr^�
s*


















zSimpleTableParser.check_columnsc
Csnd
d�|jD�
}
d}|j
rPx4tt|j�
�
D]"}|j|dd|j
kr*|}Pq*W|
|jd|�|j|d�fS)Nc
Ssg|]\}
}||
�qSrr)rCrUrXrrrrDsz:SimpleTableParser.structure_from_cells.<locals>.<listcomp>r
rY)rMr rrrO)rrHZfirst_body_rowr"rrrr
s








z&SimpleTableParser.structure_from_cells)
N)r
rrrrJrKrrQrrrPr\rRr^rrrrrrLH
s-


 
 rLcCs0x*t|
j
��
D]\}}|j|g�j|�

qWd
S)z�
    Extend the list values of `master` with those from `newdata`.

    Both parameters must be dictionaries containing list values.
    N)rF�items�
setdefaultr4)Zmaster�newdata�key�valuesrrrr0s
r0)
rZ
__docformat__rJraZdocutilsrZdocutils.utilsrrrr%rLr0rrrr�<module>s


0nR