hello @paul,

I’ve adapted your script to use the Skia font, and was able to reproduce the issue. I think it‘s related to hinting and text rasterization on low resolutions.

you can reduce the jiggle effect by converting the text to curves – that is, setting it with BezierPath instead of text().

the script below generates two movies for comparison, one with each approach.

w = 1080
h = 1080
customFont = 'Skia'
nframes = 500
fps = 30

var = listFontVariations(customFont)
varMin = var['wght']['minValue']
varMax = var['wght']['maxValue']

for mode in range(2):
    for i in range(nframes):
        newPage(w, h)
        frameDuration(1.0/fps)
        fill(1)
        rect(0, 0, w, h)
        fill(0)    
        wt = varMin + i * ((varMax - varMin) / nframes)

        # mode 0: text as text
        if mode == 0:
            font(customFont, 200)
            fontVariations(wght=wt)
            text("Hamburg", (w/2, h/2-100), align="center")
        
        # mode 1: text as bezierPath
        else:
            txt = FormattedString()
            txt.append("Hamburg", font=customFont, fontSize=200, fontVariations=dict(wght=wt))
            B = BezierPath()
            B.text(txt, (w/2, h/2-100), align="center")
            drawPath(B)
    
    saveImage(f"animation-{mode}.mp4")
    newDrawing()

hope this helps! cheers

hi. you can use a FormattedString:

txt = FormattedString()
txt.append("hello world", font='Skia', fontSize=150, fontVariations=dict(wdth=0.8, wght=1.3))

B = BezierPath()
B.text(txt, (100, 100))

drawPath(B)

hello @eduairet,

you can use the pageNumber argument in image or imageSize to get a given frame in a multipage image (gif or pdf). important: page numbering starts at 1.

see Inserting a PDF

path = '~/Desktop/image.gif'
pageCount = numberOfPages(path)

for pageNumber in range(1, pageCount+1):
    w, h = imageSize(path, pageNumber=pageNumber)
    newPage(w, h)
    image(path, (0, 0), pageNumber=pageNumber)

cheers!

hello @guidoferreyra,

ovals and rects are also available as methods of BezierPath, so you can use that same solution to draw the stroke inwards.

to draw the stroke outwards, you can draw the shape with a doubled stroke first, and then draw it again on top using only the fill (so the internal half of the stroke is covered).

here’s an updated example showing all 3 “stroke modes”:

STROKEWIDTH = 20

B = BezierPath()
B.text('a', (0, 0), fontSize=1200)
# B.rect(0, 0, 500, 500)
# B.oval(0, 0, 500, 500)

size(2200, 1000)
translate(50, 100)
fontSize(72)

for option in ['inside', 'center', 'outside']:

    text(option, (100, 780))
    save()        
    stroke(1, 0, 0)

    if option == 'inside':
        strokeWidth(STROKEWIDTH * 2)
        with savedState():
            clipPath(B)
            drawPath(B)

    elif option == 'outside':
        strokeWidth(STROKEWIDTH * 2)
        drawPath(B)
        stroke(None)
        drawPath(B)

    else:
        strokeWidth(STROKEWIDTH)
        drawPath(B)

    restore()
    translate(700, 0)

hope this helps! cheers

hello @ryan,

I dont’t know why the extremes are angled…

here’s a way to get the desired result by modifying oval:

def twist(func, angle):
    def wrapper(x, y, w, h):
        with savedState():
            translate(x + w/2, y + h/2)
            rotate(angle)
            func(-w/2, -h/2, w, h)
    return wrapper

oval = twist(oval, 45)

when working with BezierPath, it’s possible to rotate the shape:

B = BezierPath()
B.oval(x, y, w, h)
B.rotate(45, (x + w/2, y + h/2))

hope this helps!

@michelangelo you can use DrawBot inside RoboFont with the DrawBot extension

hello @RicardGarcia,

for a simple interpolation between two masters you can use font.interpolate(factor, font1, font2) – see this example script.

hope this helps!

hello @MauriceMeilleur,

assigning a list to a new variable does not create a new list, just a new reference to the same list. (this also applies to dictionaries; both are mutable object types.)

there are different ways to create a copy of a list:

A = [1, 2, 3]
B = A.copy() # or list(A) # or A[:]
B.append(4)
print(A)
print(B)

cheers!

hello @ryan,

I think newDrawing() is what you’re looking for.

cheers!

hello @MauriceMeilleur,

you can set a different resolution when saving images in DrawBot. this can produce a different size in pixels than the document size (in points):

size(100, 100)
rect(10, 10, 80, 80)
imgPath = '~/Desktop/malevich.png'
saveImage(imgPath, imageResolution=144) # 200% = 72dpi * 2
print(imageSize(imgPath))
# (200, 200)

you can also try using PIL/Pillow.

cheers!

hello @craig,

it looks like variable optical sizes are selected automatically based on the point size. you can then use scale() to make text bigger or smaller:

ttfPath = 'SourceSerif4Variable-Roman.ttf'
txt = 'Handgloves'

font(ttfPath)
translate(20, 100)

with savedState():
    scale(40)
    fontSize(6)
    text(txt, (0, 0))

translate(0, 300)
with savedState():
    scale(13)
    fontSize(19)
    text(txt, (0, 0))

translate(0, 300)
with savedState():
    scale(4)
    fontSize(72)
    text(txt, (0, 0))

hope this helps! cheers

hello @paul,

I’ve adapted your script to use the Skia font, and was able to reproduce the issue. I think it‘s related to hinting and text rasterization on low resolutions.

you can reduce the jiggle effect by converting the text to curves – that is, setting it with BezierPath instead of text().

the script below generates two movies for comparison, one with each approach.

w = 1080
h = 1080
customFont = 'Skia'
nframes = 500
fps = 30

var = listFontVariations(customFont)
varMin = var['wght']['minValue']
varMax = var['wght']['maxValue']

for mode in range(2):
    for i in range(nframes):
        newPage(w, h)
        frameDuration(1.0/fps)
        fill(1)
        rect(0, 0, w, h)
        fill(0)    
        wt = varMin + i * ((varMax - varMin) / nframes)

        # mode 0: text as text
        if mode == 0:
            font(customFont, 200)
            fontVariations(wght=wt)
            text("Hamburg", (w/2, h/2-100), align="center")
        
        # mode 1: text as bezierPath
        else:
            txt = FormattedString()
            txt.append("Hamburg", font=customFont, fontSize=200, fontVariations=dict(wght=wt))
            B = BezierPath()
            B.text(txt, (w/2, h/2-100), align="center")
            drawPath(B)
    
    saveImage(f"animation-{mode}.mp4")
    newDrawing()

hope this helps! cheers

hello @eduairet,

when working with QRCodeGenerator:

• data must be in bytes
• size must be a tuple

this works:

path = bytes('http://www.drawbot.com/', 'utf-8')

w, h = 31, 31

img = ImageObject()
img.QRCodeGenerator((w, h), path, 'Q')

size(w, h)
image(img, (0, 0))

the third parameter (L, M, Q or H) determines the correction level, see the Core Image docs for more info.

hope this helps! cheers

ps. the resulting image is tiny (1 pixel per bit). you can scale it by reading the pixel values and redrawing it with shapes:

path = bytes('http://www.drawbot.com/', 'utf-8')

w, h = 31, 31
s = 10

img = ImageObject()
img.QRCodeGenerator((w, h), path, 'Q')

size(w*s, h*s)
for x in range(w):
    for y in range(h):
        c = imagePixelColor(img, (x, y))
        fill(*c)
        rect(x*s, y*s, s, s)

hello @djrrb,

I think this error was introduced in markdown 3.3 (new htmlparser).

I don’t know how to fix it, but you can install an earlier version of markdown which still works:

pip install markdown==3.2.2

cheers

@imik glad to hear that. cheers!!

hello @snaders,

yes, you can read image pixel values with imagePixelColor(). it’s a bit slow though, use it with caution… good luck!

txt = "Boustrophedon is a type of bi-directional text, mostly seen in ancient manuscripts and other inscriptions. Alternate lines of writing are flipped, or reversed, with reversed letters. Rather than going left-to-right as in modern European languages, or right-to-left as in Arabic and Hebrew, alternate lines in boustrophedon must be read in opposite directions. Also, the individual characters are reversed, or mirrored. It was a common way of writing in stone in ancient Greece.".upper()


def boustrophedonicTextBox(txt, box, drawFrame=False):

    x, y, w, h = box
    L = fontLineHeight()

    save()
    translate(x, y-L)

    T = ''
    lineLength = 0
    backward = False
    words = txt.split()
    lineCount = 0
    overflow = ''

    for i, word in enumerate(words):

        # add word to current line
        if lineLength + textSize(word)[0] <= w:
            T += word + ' '
            lineLength += textSize(word + ' ')[0]

        # break to next line
        else:
            # no more space for new line
            if not (lineCount+1) * L < h:
                overflow = T + ' '.join(words[i:])
                break

            # draw current line
            with savedState():
                # flip direction
                if backward:
                    scale(-1, 1)
                text(T, (0, 0), align='center')

            # begin new line
            translate(0, -L)
            backward = not backward
            lineCount += 1
            T = word + ' '
            lineLength = textSize(word + ' ')[0]

        # draw last line
        if i == len(words)-1:
            lineCount += 1
            with savedState():
                if backward:
                    scale(-1, 1)
                text(T, (0, 0), align='center')

    restore()

    if drawFrame:
        with savedState():
            fill(None)
            stroke(1, 0, 0)
            rect(x - w/2, y, w, -h)

    return overflow


font('Skia')
fontSize(36)
lineHeight(66)
txt = boustrophedonicTextBox(txt, (width()/2, height()-100, width()-80, 800), drawFrame=False)

print(txt)

hello @mït,

I think I understand what you mean – it’s not so easy because, once a piece of text is drawn into the page, you can’t really ‘search’ for it anymore.

it can be done if you lay out the text yourself: word by word, line by line, page by page. the example script below is based on this one.

TXT = "Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum."
FONT = 'Menlo-Bold'
FONTSIZE = 36
LINEHEIGHT = 120
WIDTH = 570
X, Y = 150, 900
HIGHLIGHT = 'voluptate'
PAGEHIGHLIGHT = None

# start with an empty string, add words one by one
T = FormattedString(font=FONT, fontSize=FONTSIZE, lineHeight=FONTSIZE*1.2)

# set text properties globally - needed for correct measurements with `textSize`
font(FONT)
fontSize(FONTSIZE)

# variables to keep track of the current line
lineLength = 0
lineNumber = 1
lineY = Y

# iterate over words in text
for word in TXT.split():

    # if word fits in the current line, add word to line
    if lineLength + textSize(word)[0] <= WIDTH:
        T += word + ' '
        lineLength += textSize(word + ' ')[0]

    # word does not fit in current line, make a new line
    else:        

        # if next line does not fit in current page, start a new page
        if lineY - LINEHEIGHT < 0:
            newPage()
            font(FONT)
            fontSize(FONTSIZE)
            lineY = Y
            
        # draw the current line
        text(T, (X, lineY))

        # check if highlight word was in this line and page
        if HIGHLIGHT in str(T):
            PAGEHIGHLIGHT = pageCount()
            color = (1, 0, 0)
        else: 
            color = (0.7,)
        fill(*color)
        text(str(lineNumber), (X-50, lineY), align='right')

        # make a fresh new line, add word to line
        T = FormattedString(font=FONT, fontSize=FONTSIZE, lineHeight=FONTSIZE*1.2)
        T += word + ' '

        # update variables for new line
        lineLength = textSize(word + ' ')[0]
        lineNumber += 1
        lineY -= LINEHEIGHT

# draw page numbers, check if highlight word in page
for i, page in enumerate(pages()):
    with page:
        color = (1, 0, 0) if (i+1) == PAGEHIGHLIGHT else (0.7,)
        fill(*color)
        text(str(i+1), (width()-50, Y), align='right')

hope this makes sense!

hello @mït,

not sure if we can call it clean but here’s a way to apply formatting to a FormattedString a posteriori using a simple find & replace:

txt = '''Some title here
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nulla tempus justo in finibus condimentum. Proin vitae mauris ac nibh elementum consectetur. Aenean justo mi, cursus vel placerat vitae, gravida in mi.
'''

def applyFormatting(formattedString, findText, styleDict):

    start = str(formattedString).find(findText)
    end = start + len(findText)
    
    before = formattedString[:start]
    after  = formattedString[end:]

    replaceText = formattedString.copy()
    replaceText.clear()
    replaceText.append(findText, **styleDict)

    return before + replaceText + after

T = FormattedString(txt, fontSize=36, font="Arial")

T = applyFormatting(T, 'Some title here', dict(fontSize=72, font='Times New Roman', paragraphBottomSpacing=20))
T = applyFormatting(T, 'sit amet', dict(font='Arial Bold', fill=(1, 0, 0),))
T = applyFormatting(T, 'condimentum', dict(font='Arial Italic', fill=(0, 1, 0),))

while len(T):
    newPage('A4')
    T = textBox(T, (20, 20, width() - 40, height() - 40))

adding support for markdown or html markup to FormattedString would be super useful!

hope this helps

hello @mït,

you can change text properties in the FormattedString, before passing it to textBox.

like this:

T = FormattedString(fontSize=42)
T.font("Arial")
T.append('Lorem ipsum dolor ')

T.font("Arial Bold")
T.append('sit amet, ')

T.font("Arial")
T.append('consectetur adipiscing elit. Nulla tempus justo in finibus ')

T.font("Arial Italic")
T.append('condimentum. ')

T.font("Arial")
T.append('Proin vitae mauris ac nibh elementum consectetur. Aenean justo mi, cursus vel placerat vitae, gravida in mi.')

while len(T):
    newPage('A4')
    T = textBox(T, (10, 10, width() - 20, height() - 20))

hope this helps! cheers