inkscape_contributed/eggbot_spiraltext.py

# coding=utf-8
# eggbot_spiraltext.py
#
# Render a passage of text using the Hershey fonts, then stretch it so
# that it will wrap multiple times around an egg, and finally tilt it
# so that it will spiral as it wraps.
#
# + The wrapping need not be an integral multiple of 3200 pixels
#
# + The text tilt is computed to use the full height of the document
#
# + The text can be run starting from the top of the page or from
#   the bottom (and upside down).  This latter orientation is useful
#   when placing the bottom of the egg (fat end) in the egg motor's
#   egg cup
#
# + The text can be stretched more horizontally than vertically to
#   compensate for some of the geometry issues associated with drawing
#   on eggs.
#
# + The text can contain markup (see below)
#
# This extension also permits some basic markup of the passage using
# XHTML-like conventions and a limited set of tags:
#
#  <sans>   - A simple typeface which lacks serifs
#  <times>  - "Times" like typeface (a face with serifs)
#  <script> - A flowing script font
#  <b>      - Boldface
#  <em>     - Emphasis
#  <i>      - Italics
#  <face>   - Where "face" is any of the typeface names from hersheydata.py
#
# The markup processing is not XML-conformant: we don't expect a well-formed
# document as input.  No single root element is required.  And, at the end of
# the text, we do not require closure of any open tags.  We do however enforce
# proper nesting of tags: an element cannot be closed unless its children have
# already been closed.  This is more to prevent ambiguity about whether or
# not closing a typeface also closes any markup operating under it (e.g.,
# does <sans><b>text</sans> mean that the <b> was implicitly ended when
# </sans> was encountered?
#
# This extension requires the hersheydata.py file which is part of the
# Hershey Text rendering extension written by Windell H. Oskay of
# www.evilmadscientist.com.  Information on that extension may be found at
#
#   http://www.evilmadscientist.com/go/hershey
#
# Copyright 2011, Daniel C. Newman,
#
# Significant portions of this code were written by Windell H. Oskay and are
# Copyright 2011, Windell H. Oskay, www.evilmadscientist.com
#
# Small portions of this code were changed by Sheldon B. Michaels 2016,
# in order to accommodate the addition of several new faces
# (the "EMS" series) to hersheydata.py.  Additionally, changes were made
# to the default text rendering style.
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

import math

import hersheydata  # data file w/ Hershey font data
import inkex
import simplestyle

# Mapping table to map the names used here to the corresponding
# names used in hersheydata.py.  This helps prevent end users from
# being impacted by a name change in hersheydata.py.  This can also
# be used to deal with a face being removed from hersheydata.py

map_our_names_to_hersheydata = {
    'astrology': 'astrology',
    'cursive': 'cursive',
    'cyrillic': 'cyrillic',
    'futural': 'futural',
    'futuram': 'futuram',
    'gothiceng': 'gothiceng',
    'gothicger': 'gothicger',
    'gothicita': 'gothicita',
    'greek': 'greek',
    'japanese': 'japanese',
    'markers': 'markers',
    'mathlow': 'mathlow',
    'mathupp': 'mathupp',
    'meteorology': 'meteorology',
    'music': 'music',
    'scriptc': 'scriptc',
    'scripts': 'scripts',
    'symbolic': 'symbolic',
    'timesg': 'timesg',
    'timesi': 'timesi',
    'timesib': 'timesib',
    'timesr': 'timesr',
    'timesrb': 'timesrb'}


# The following two routines are lifted with impunity from Windell H. Oskay's
# hershey.py Hershey Text extension for Inkscape.  They are,
# Copyright 2011, Windell H. Oskay, www.evilmadscientist.com

def draw_svg_text(char, face, offset, vertoffset, parent):
    style = {'stroke': '#000000', 'fill': 'none', 'stroke-linecap': 'round', 'stroke-linejoin': 'round'}
    # Apply rounding to ends so that user gets best impression of final printed text appearance.
    path_string = face[char]
    split_string = path_string.split()
    midpoint = offset - float(split_string[0])
    i = path_string.find("M")
    if i >= 0:
        path_string = path_string[i:]  # portion after first move
        trans = 'translate(' + str(midpoint) + ',' + str(vertoffset) + ')'
        text_attribs = {'style': simplestyle.formatStyle(style), 'd': path_string, 'transform': trans}
        inkex.etree.SubElement(parent, inkex.addNS('path', 'svg'), text_attribs)
    return midpoint + float(split_string[1])  # new offset value


def renderText(parent, markup):
    # Embed text in group to make manipulation easier:
    g_attribs = {inkex.addNS('label', 'inkscape'): 'Hershey Text'}
    g = inkex.etree.SubElement(parent, 'g', g_attribs)

    w = 0  # Initial spacing offset
    spacing = 3  # spacing between letters

    for face, text in markup:
        if face in map_our_names_to_hersheydata:
            face = map_our_names_to_hersheydata[face]
        font = getattr(hersheydata, face)
        letter_vals = (ord(q) - 32 for q in text)
        for q in letter_vals:
            if (q < 0) or (q > 95):
                w += 2 * spacing
            else:
                w = draw_svg_text(q, font, w, 0, g)
    return g, w


# The generic font "families" we support
generic_families = ('sans', 'script', 'times')

# Convert "family-name" + "bold-0-or-1" + "italic-0-or-1" to a typeface name
family_to_font = {'sans00': 'futural',
                  'sans10': 'futuram',
                  'sans01': 'futural',
                  'sans11': 'futuram',
                  'times00': 'timesr',
                  'times10': 'timesrb',
                  'times01': 'timesi',
                  'times11': 'timesib',
                  'script00': 'scripts',
                  'script10': 'scriptc',
                  'script01': 'scripts',
                  'script11': 'scriptc'}
emphasis_is_bold = {'sans': True, 'times': False, 'script': True}

# Short list of entity references
entity_refs = {'&lt;': '<', '&gt;': '>', '&amp;': '&', '&quot;': '"', '&apos': "'", '&nbsp;': ' '}


def normalize_possible_EMS_string(tag):
    # Normalizes tag name by removing any spaces
    s_normalized_tag = tag.replace(' ', '')
    return s_normalized_tag


def is_valid_EMS_name(tag):
    # returns true if family is one of the "EMS" faces in hersheydata.py
    # else false
    s_normalized_tag = normalize_possible_EMS_string(tag)
    b_ret_val = False  # default assumption
    try:
        fontgroup = hersheydata.group_allfonts
    except:
        # User probably has old version of hersheydata.py
        pass
    else:
        for f in fontgroup:
            if f[0] == s_normalized_tag:
                b_ret_val = True
                break

    return b_ret_val


def pickFace(family, bold=False, italics=False, emphasis=False):
    if not family:
        return None

    b = '0'
    i = '0'

    # If using a generic font family, then determine how to map <em>
    if emphasis and (family in generic_families):
        if emphasis_is_bold[family]:
            bold = True
        else:
            italics = True

    if bold:
        b = '1'

    if italics:
        i = '1'

    if (family + b + i) in family_to_font:
        return family_to_font[family + b + i]

    return family


def processMarkup(text, family='sans'):
    if text is None:
        text = ''

    # By default we assume 'sans'
    if not family:
        family = 'sans'
    default_family = family
    face_stack = [family]

    # Bold and italics off
    bold = False
    emphasis = False
    italic = False

    # Set the current typeface
    face = pickFace(family, bold, italic, emphasis)

    # And the result of markup processing so far
    markup = []

    # We keep a queue / list of the open markup tags
    # When a tag is closed, we expect it to be well nested.  To enforce
    # that expectation, we make sure that we are closing the most recently
    # opened tag.  While this may seem overly picky, it's easier than worrying
    # issues like, "Does closing a typeface imply implicitly closing <b> or <it>?"
    # And, "Does starting a new typeface imply closing the prior selected face?"
    tags_used = []

    outstr = ''
    i = 0
    while i < len(text):
        # An entity reference?
        if text[i] == '&':
            j = text.find(';', i + 1)
            if j != -1:
                eref = text[i:j + 1]
                if eref in entity_refs:
                    outstr += entity_refs[eref]
                    i = j + 1
                else:
                    inkex.errormsg('Ignoring the unrecognized entity reference {0}.'.format(eref))
                    outstr += eref
                    i = j + 1
            else:
                inkex.errormsg('An unescaped "&" was encountered; please replace it with "&amp;".')
                break

        # Start of a tag (start-tag or end-tag? self-closing tags not supported)
        elif text[i] == '<':
            j = text.find('>', i + 1)
            if (j != -1) and (j > (i + 1)):

                tag = text[i + 1:j]
                i = j + 1

                if tag[0] == '/':
                    # This is an end-tag (closing tag)
                    close = True
                    tag = tag[1:]

                    # Ensure that the most recently opened tag is that which we are closing here
                    # We'll pop the most recent tag from the queue of opened tags and see if
                    # it matches
                    if len(tags_used) == 0:
                        inkex.errormsg('The ending tag </{0}> appeared before any start tag <{1}>.'.format(tag, tag))
                        break
                    else:
                        old_tag = tags_used.pop()
                        if old_tag != tag:
                            inkex.errormsg('The ending tag </{0}> does not appear to be correctly nested; it tried to close the tag <{1}>.  Sorry, but all tags must be properly nested.'.format(tag, old_tag))
                            break
                else:
                    # Start tag (opening tag)
                    # Push it onto the queue of opened tags
                    close = False
                    tags_used.append(tag)

                if (tag == 'b') or (tag == 'strong'):
                    if bold == close:
                        # Push prior string and font onto the stack
                        if outstr != '':
                            markup.append([face, outstr])
                            outstr = ''

                        # Start a new boldface string
                        bold = not bold
                        face = pickFace(family, bold, italic, emphasis)

                elif tag == 'i':
                    if italic == close:
                        # Push the prior string and font unto the stack
                        if outstr != '':
                            markup.append([face, outstr])
                            outstr = ''

                        # Start a new italicized string
                        italic = not italic
                        face = pickFace(family, bold, italic, emphasis)

                elif tag == 'em':
                    if emphasis == close:
                        # Push the prior string and font unto the stack
                        if outstr != '':
                            markup.append([face, outstr])
                            outstr = ''

                        # Start a new italicized string
                        emphasis = not emphasis
                        face = pickFace(family, bold, italic, emphasis)

                else:
                    b_valid_ems_name = is_valid_EMS_name(tag)
                    if b_valid_ems_name:
                        tag = normalize_possible_EMS_string(tag)
                    if all([tag not in generic_families,
                            tag not in map_our_names_to_hersheydata,
                            not b_valid_ems_name]):
                        if close:
                            inkex.errormsg('Ignoring the unrecognized tag </{0}>.'.format(tag))
                        else:
                            inkex.errormsg('Ignoring the unrecognized tag <{0}>.'.format(tag))
                    else:
                        if outstr != '':
                            markup.append([face, outstr])
                            outstr = ''
                        if not close:
                            family = tag
                            face_stack.append(family)
                        else:
                            if face_stack:
                                # Current face on the stack should be the one we just closed
                                face_stack.pop()
                                if face_stack:
                                    family = face_stack[len(face_stack) - 1]
                                else:
                                    family = default_family
                            else:
                                family = default_family
                        face = pickFace(family, bold, italic, emphasis)
            else:
                inkex.errormsg('Ignoring unescaped "<"')
                outstr += '<'
                i += 1
        else:
            outstr += text[i]
            i += 1

    # We won't worry about unclosed tags -- we're not trying to be an XML or XHTML parser

    # See if there was a hard error
    if i < len(text):
        return None

    # And push the last text into the list of processed markup
    if outstr != '':
        markup.append([face, outstr])

    return markup


class SpiralText(inkex.Effect):

    def __init__(self):
        inkex.Effect.__init__(self)
        self.OptionParser.add_option("--tab",  # NOTE: value is not used.
                                     action="store", type="string",
                                     dest="tab", default="splash",
                                     help="The active tab when Apply was pressed")
        self.OptionParser.add_option("--text",
                                     action="store", type="string",
                                     dest="text", default="Hershey Text for Inkscape",
                                     help="The input text to render")
        self.OptionParser.add_option("--fontfamily",
                                     action="store", type="string",
                                     dest="fontfamily", default="sans",
                                     help="The selected font face when Apply was pressed")
        self.OptionParser.add_option("--wrap",
                                     action="store", type="float",
                                     dest="wrap", default=10.0,
                                     help="Number of times to wrap the text around the egg")
        self.OptionParser.add_option("--flip",
                                     action="store", type="inkbool",
                                     dest="flip", default=False,
                                     help="Flip the text for plotting with the egg's bottom at the egg motor")
        self.OptionParser.add_option("--stretch",
                                     action="store", type="inkbool",
                                     dest="stretch", default=True,
                                     help="Stretch the text horizontally to account for egg distortions")

    def effect(self):

        markup = processMarkup(self.options.text, self.options.fontfamily)
        g, w = renderText(self.current_layer, markup)

        # Now to wrap the text N times around the egg, we need to scale it to have
        # length 3200 * N.  It's current width is w so the scale factor is (3200 * N) / w.

        scale_x = float(3200 * self.options.wrap) / float(w)
        scale_y = scale_x
        if self.options.stretch:
            scale_y = scale_y * 2.0 / 3.0

        # In planning the scaling, we'd like to know the height of our line of text.
        # Rather than computing its bounding box, we'll just use the height of the
        # parens from the Simplex Roman font.  And, we could compute that but we'll
        # just use our prior knowledge of it being 32.

        h = 32.0

        # And the angular tilt will be arcsine( height / (3200 * fWrap) )
        svg = self.document.getroot()
        height = float(self.unittouu(svg.attrib['height'])) - h * scale_y
        angle = (180.0 / math.pi) * math.asin(height / float(3200 * self.options.wrap))

        if self.options.flip:
            angle += 180.0
            t = 'translate({0:f},{1:f}) rotate({2:f},{3:f},0) scale({4:f},{5:f})'.format(-w * scale_x, h * scale_y, angle,
                                                                                   w * scale_x, scale_x, scale_y)
        else:
            t = 'translate(0,{0:f}) rotate({1:f},0,0) scale({2:f},{3:f})'.format(h, angle, scale_x, scale_y)
        g.set('transform', t)


if __name__ == '__main__':
    e = SpiralText()
    e.affect()