We're using iText to put a text inside a signature placeholder in a PDF. We use a code snippet similar to this to define the Signature Appearence
PdfStamper stp = PdfStamper.createSignature(inputReader, os, '\0', tempFile2, true);
sap = stp.getSignatureAppearance();
sap.setVisibleSignature(placeholder);
sap.setRenderingMode(PdfSignatureAppearance.RenderingMode.DESCRIPTION);
sap.setCertificationLevel(PdfSignatureAppearance.NOT_CERTIFIED);
Calendar cal = Calendar.getInstance();
sap.setSignDate(cal);
sap.setLayer2Text(text+"\n"+cal.getTime().toString());
sap.setReason(text+"\n"+cal.getTime().toString()); `
Everything works fine, but the signature text does not fill all the signature placeholder area as expected by us, but the area filled seems to have an height that is approximately the 70% of the available space.
As a result, sometimes especially if the length of the signature text is quite big, the signature text does not fit in the placeholder and the text is striped away.
Example of filled Signature:
I looked into the PdfSignatureAppearence class and I found this code snippet in the getApperance() method that is responsible of this behaviour and is invoked when
sap.setRenderingMode(PdfSignatureAppearance.RenderingMode.DESCRIPTION);
is being called
else {
dataRect = new Rectangle(
MARGIN,
MARGIN,
rect.getWidth() - MARGIN,
rect.getHeight() * (1 - TOP_SECTION) - MARGIN);
}
I don't get the reason for that, because I expect that the text could use all the available placeholder height, with the proper margin.
Is there any way to bypass this behaviour?
We are using iText 5.4.2, but also newer version contains same code snippet so I expect that the behaviour will be same.
As #JJ. already commented,
TOP_SECTION is connected with acro6layers rendering and the code [determining the datarect in pure DESCRIPTION mode] does not take into account the value of the acro6layer flag.
Unless one wants to fix this in the iText 5 code itself, the easiest way to make one's description use the whole signature space is to construct the layer 2 appearance oneself.
To do so one merely has to retrieve a PdfTemplate from PdfSignatureAppearance.getLayer(2) and fill it as desired after one has called PdfSignatureAppearance.setVisibleSignature. The PdfSignatureAppearance remembers that you already have retrieved the layer 2 and doesn't change it anymore.
For the case at hand we essentially copy the PdfSignatureAppearance.getAppearance code for generating layer 2 in pure DESCRIPTION mode, merely correcting the code determining the datarect:
PdfSignatureAppearance appearance = ...;
[...]
appearance.setVisibleSignature(new Rectangle(36, 748, 144, 780), 1, "sig");
PdfTemplate layer2 = appearance.getLayer(2);
String text = "We're using iText to put a text inside a signature placeholder in a PDF. "
+ "We use a code snippet similar to this to define the Signature Appearence.\n"
+ "Everything works fine, but the signature text does not fill all the signature "
+ "placeholder area as expected by us, but the area filled seems to have an height "
+ "that is approximately the 70% of the available space.\n"
+ "As a result, sometimes especially if the length of the signature text is quite "
+ "big, the signature text does not fit in the placeholder and the text is striped "
+ "away.";
Font font = new Font();
float size = font.getSize();
final float MARGIN = 2;
Rectangle dataRect = new Rectangle(
MARGIN,
MARGIN,
appearance.getRect().getWidth() - MARGIN,
appearance.getRect().getHeight() - MARGIN);
if (size <= 0) {
Rectangle sr = new Rectangle(dataRect.getWidth(), dataRect.getHeight());
size = ColumnText.fitText(font, text, sr, 12, appearance.getRunDirection());
}
ColumnText ct = new ColumnText(layer2);
ct.setRunDirection(appearance.getRunDirection());
ct.setSimpleColumn(new Phrase(text, font), dataRect.getLeft(), dataRect.getBottom(), dataRect.getRight(), dataRect.getTop(), size, Element.ALIGN_LEFT);
ct.go();
(CreateSignature.java test signWithCustomLayer2)
(As description text I used some paragraphs from the question body.)
The result:
By adapting the MARGIN value in the code above, one can even use more are. As that can result in the text touching the border, though, that might not be really beautiful.
As an aside:
if the length of the signature text is quite big, the signature text does not fit in the placeholder and the text is striped away.
If you initialize the size variable above with a non-positive value, the code in the if (size <= 0) block will calculate a font size which allows all of the text to fit into the signature rectangle. This does happen in the code above as new Font() returns a font with a size of UNDEFINED which is a constant -1.
When I parse an existing PDF using iText(Sharp), I create an object which implements IRenderListener which I pass into PdfReaderContentParser.ProcessContent() and sure enough, my object's RenderText() gets called repeatedly with all the text in the PDF.
The problem is, the TextRenderInfo tells me about the base font (in my case, Helvetica) but I can't tell the height of the font nor its weight (regular vs. bold). Is this a known deficiency of iText(Sharp) or am I missing something?
the TextRenderInfo tells me about the base font (in my case, Helvetica) but I can't tell the height of the font nor its weight (regular vs. bold)
Height
Unfortunately iTextSharp does not provide a public font size method or member in the TextRenderInfo. Some people worked around this by using the distance between its GetAscentLine() and its GetDescentLine().
If you are ready to use Reflection, though, you can do better by exposing and using the private TextRenderInfo member GraphicsState gs, e.g. like in this render listener:
public class LocationTextSizeExtractionStrategy : LocationTextExtractionStrategy
{
//Hold each coordinate
public List<SizeAndTextAndFont> myChunks = new List<SizeAndTextAndFont>();
//Automatically called for each chunk of text in the PDF
public override void RenderText(TextRenderInfo wholeRenderInfo)
{
base.RenderText(wholeRenderInfo);
GraphicsState gs = (GraphicsState) GsField.GetValue(wholeRenderInfo);
myChunks.Add(new SizeAndTextAndFont(gs.FontSize, wholeRenderInfo.GetText(), wholeRenderInfo.GetFont().PostscriptFontName));
}
FieldInfo GsField = typeof(TextRenderInfo).GetField("gs", System.Reflection.BindingFlags.NonPublic | System.Reflection.BindingFlags.Instance);
}
//Helper class that stores our rectangle, text, and font
public class SizeAndTextAndFont
{
public float Size;
public String Text;
public String Font;
public SizeAndTextAndFont(float size, String text, String font)
{
this.Size = size;
this.Text = text;
this.Font = font;
}
}
You can extract information with such a render listener like this:
using (var pdfReader = new PdfReader(testFile))
{
// Loop through each page of the document
for (var page = startPage; page < endPage; page++)
{
Console.WriteLine("\n Page {0}", page);
LocationTextSizeExtractionStrategy strategy = new LocationTextSizeExtractionStrategy();
PdfTextExtractor.GetTextFromPage(pdfReader, page, strategy);
foreach (SizeAndTextAndFont p in strategy.myChunks)
{
Console.WriteLine(string.Format("<{0}> in {2} at {1}", p.Text, p.Size, p.Font));
}
}
}
This produces an output like this:
Page 1
< The Philippine Stock Exchange, Inc> in Helvetica-Bold at 8
< Daily Quotations Report> in Helvetica-Bold at 8
< March 23 , 2015> in Helvetica-Bold at 8
<Name> in Helvetica at 7
<Symbol> in Helvetica at 7
<Bid> in Helvetica at 7
[...]
Considering transformations
The numbers you see in the output as font sizes are the values of the font size property in the PDF graphics state at the time the respective text is drawn.
Due to the flexibility of PDF this may not be font size you eventually see in the output, though, a custom transformation may stretch the output considerably. Some PDF producers even always use a font size of 1 and transformations to stretch the output accordingly.
To get a good value for font sizes in such documents, you can improve the LocationTextSizeExtractionStrategy method RenderText like this:
public override void RenderText(TextRenderInfo wholeRenderInfo)
{
base.RenderText(wholeRenderInfo);
GraphicsState gs = (GraphicsState) GsField.GetValue(wholeRenderInfo);
Matrix textToUserSpaceTransformMatrix = (Matrix) TextToUserSpaceTransformMatrixField.GetValue(wholeRenderInfo);
float transformedFontSize = new Vector(0, gs.FontSize, 0).Cross(textToUserSpaceTransformMatrix).Length;
myChunks.Add(new SizeAndTextAndFont(transformedFontSize, wholeRenderInfo.GetText(), wholeRenderInfo.GetFont().PostscriptFontName));
}
with this additional reflection FieldInfo member.
FieldInfo TextToUserSpaceTransformMatrixField = typeof(TextRenderInfo).GetField("textToUserSpaceTransformMatrix", System.Reflection.BindingFlags.NonPublic | System.Reflection.BindingFlags.Instance);
Weight
As you can see in the output above, the name of the font may contain more than the mere font family name but also a weight indicator
< March 23 , 2015> in Helvetica-Bold at 8
In your example, therefore,
the TextRenderInfo tells me about the base font (in my case, Helvetica)
the Helvetica without any decorations would imply a regular weight.
Helvetica is one of the standard 14 fonts which every PDF viewer must provide out-of-the-box: Times-Roman, Helvetica, Courier, Symbol, Times-Bold, Helvetica-Bold, Courier-Bold, ZapfDingbats, Times-Italic, Helvetica-Oblique, Courier-Oblique, Times-BoldItalic, Helvetica-BoldOblique, Courier-BoldOblique. Thus, these names are pretty dependable.
Unfortunately font names in general may be chosen arbitrarily; a bold font may have "Bold" or "Black" or other indicators of boldness in its name or none at all.
One might also try to use the font's FontDescriptor dictionary for which an entry FontWeight is specified. Unfortunately this entry is optional, you cannot count on it being there at all.
Furthermore, a font in a PDF can be artificially bold'ed, cf. this answer:
All these numbers are drawn using the same font, merely adding a rising outline line width.
Thus, I'm afraid there is no dependable way to find the exact font weight, merely a number of heuristics which may or may not return acceptable approximations.
I have a PDF where I add some TextFields.
var txtFld = new TextField(stamper.Writer, new Rectangle(cRightX - cWidthX, cTopY3, cRightX, cTopY), FieldNameProtocol) { Font = bf, FontSize = cHeaderFontSize, Alignment = Element.ALIGN_RIGHT, Options = PdfFormField.FF_MULTILINE };
stamper.AddAnnotation(txtFld.GetTextField(), 1);
The ‘bf’ above is a Unicode font that gets embedded in the PDF:
BaseFont bf = BaseFont.CreateFont(UnicodeFontPath, BaseFont.IDENTITY_H, BaseFont.EMBEDDED); // Create a Unicode font to write in Greek...
Later-on I fill those fields with greek text.
var acrof = stamper.AcroFields;
acrof.SetField(fieldName, field.Value/*, field.Value*/); // Set the text of the form field.
acrof.SetFieldProperty(fieldName, "setfflags", PdfFormField.FF_READ_ONLY, null); // Make it readonly.
When I view the PDF, most of the times the text is missing and if I click on the (invisible) TextField in Acrobat, then the text becomes visible (until it loses focus again).
Any idea what is going on here?
I have also tried using non-embedded font, but I get the same thing (although I still seem to get embedded fonts in PDF that are similar to the font I use). I don't know if I am missing sth.
It seemed that I was doing the following at the wrong order (the following is the correct):
acrof.SetFieldProperty(field.Name, "setfflags", PdfFormField.FF_READ_ONLY, null); // Make it readonly.
acrof.SetFieldProperty(field.Name, "textfont", bf, null);
acrof.SetField(field.Name, field.Value/*, field.Value*/); // Set the text of the form field.
At least that's hat I think it was wrong.
I have made many changes.
I'm using an NSMutableAttribtuedString in order to build a string with formatting, which I then pass to Core Text to render into a frame. The problem is, that I need to use superscript and subscript. Unless these characters are available in the font (most fonts don't support it), then setting the property kCTSuperscriptAttributeName does nothing at all.
So I guess I'm left with the only option, which is to fake it by changing the font size and moving the base line. I can do the font size bit, but don't know the code for altering the base line. Can anyone help please?
Thanks!
EDIT: I'm thinking, considering the amount of time I have available to sort this problem, of editing a font so that it's given a subscript "2"... Either that or finding a built-in iPad font which does. Does anyone know of any serif font with a subscript "2" I can use?
There is no baseline setting amongst the CTParagraphStyleSpecifiers or the defined string attribute name constants. I think it's therefore safe to conclude that CoreText does not itself support a baseline adjust property on text. There's a reference made to baseline placement in CTTypesetter, but I can't tie that to any ability to vary the baseline over the course of a line in the iPad's CoreText.
Hence, you probably need to interfere in the rendering process yourself. For example:
create a CTFramesetter, e.g. via CTFramesetterCreateWithAttributedString
get a CTFrame from that via CTFramesetterCreateFrame
use CTFrameGetLineOrigins and CTFrameGetLines to get an array of CTLines and where they should be drawn (ie, the text with suitable paragraph/line breaks and all your other kerning/leading/other positioning text attributes applied)
from those, for lines with no superscript or subscript, just use CTLineDraw and forget about it
for those with superscript or subscript, use CTLineGetGlyphRuns to get an array of CTRun objects describing the various glyphs on the line
on each run, use CTRunGetStringIndices to determine which source characters are in the run; if none that you want to superscript or subscript are included, just use CTRunDraw to draw the thing
otherwise, use CTRunGetGlyphs to break the run into individual glyphs and CTRunGetPositions to figure out where they would be drawn in the normal run of things
use CGContextShowGlyphsAtPoint as appropriate, having tweaked the text matrix for those you want in superscript or subscript
I haven't yet found a way to query whether a font has the relevant hints for automatic superscript/subscript generation, which makes things a bit tricky. If you're desperate and don't have a solution to that, it's probably easier just not to use CoreText's stuff at all — in which case you should probably define your own attribute (that's why [NS/CF]AttributedString allow arbitrary attributes to be applied, identified by string name) and use the normal NSString searching methods to identify regions that need to be printed in superscript or subscript from blind.
For performance reasons, binary search is probably the way to go on searching all lines, the runs within a line and the glyphs within a run for those you're interested in. Assuming you have a custom UIView subclass to draw CoreText content, it's probably smarter to do it ahead of time rather than upon every drawRect: (or the equivalent methods, if e.g. you're using a CATiledLayer).
Also, the CTRun methods have variants that request a pointer to a C array containing the things you're asking for copies of, possibly saving you a copy operation but not necessarily succeeding. Check the documentation. I've just made sure that I'm sketching a workable solution rather than necessarily plotting the absolutely optimal route through the CoreText API.
Here is some code based on Tommy's outline that does the job quite well (tested on only single lines though). Set the baseline on your attributed string with #"MDBaselineAdjust", and this code draws the line to offset, a CGPoint. To get superscript, also lower the font size a notch. Preview of what's possible: http://cloud.mochidev.com/IfPF (the line that reads "[Xe] 4f14...")
Hope this helps :)
NSAttributedString *string = ...;
CGPoint origin = ...;
CTFramesetterRef framesetter = CTFramesetterCreateWithAttributedString((CFAttributedStringRef)string);
CGSize suggestedSize = CTFramesetterSuggestFrameSizeWithConstraints(framesetter, CFRangeMake(0, string.length), NULL, CGSizeMake(CGFLOAT_MAX, CGFLOAT_MAX), NULL);
CGPathRef path = CGPathCreateWithRect(CGRectMake(origin.x, origin.y, suggestedSize.width, suggestedSize.height), NULL);
CTFrameRef frame = CTFramesetterCreateFrame(framesetter, CFRangeMake(0, string.length), path, NULL);
NSArray *lines = (NSArray *)CTFrameGetLines(frame);
if (lines.count) {
CGPoint *lineOrigins = malloc(lines.count * sizeof(CGPoint));
CTFrameGetLineOrigins(frame, CFRangeMake(0, lines.count), lineOrigins);
int i = 0;
for (id aLine in lines) {
NSArray *glyphRuns = (NSArray *)CTLineGetGlyphRuns((CTLineRef)aLine);
CGFloat width = origin.x+lineOrigins[i].x-lineOrigins[0].x;
for (id run in glyphRuns) {
CFRange range = CTRunGetStringRange((CTRunRef)run);
NSDictionary *dict = [string attributesAtIndex:range.location effectiveRange:NULL];
CGFloat baselineAdjust = [[dict objectForKey:#"MDBaselineAdjust"] doubleValue];
CGContextSetTextPosition(context, width, origin.y+baselineAdjust);
CTRunDraw((CTRunRef)run, context, CFRangeMake(0, 0));
}
i++;
}
free(lineOrigins);
}
CFRelease(frame);
CGPathRelease(path);
CFRelease(framesetter);
`
You can mimic subscripts now using TextKit in iOS7. Example:
NSMutableAttributedString *carbonDioxide = [[NSMutableAttributedString alloc] initWithString:#"CO2"];
[carbonDioxide addAttribute:NSFontAttributeName value:[UIFont systemFontOfSize:8] range:NSMakeRange(2, 1)];
[carbonDioxide addAttribute:NSBaselineOffsetAttributeName value:#(-2) range:NSMakeRange(2, 1)];
I've been having trouble with this myself. Apple's Core Text documentation claims that there has been support in iOS since version 3.2, but for some reason it still just doesn't work. Even in iOS 5... how very frustrating >.<
I managed to find a workaround if you only really care about superscript or subscript numbers. Say you have a block of text can might contain a "sub2" tag where you want a subscript number 2. Use NSRegularExpression to find the tags, and then use replacementStringForResult method on your regex object to replace each tag with unicode characters:
if ([match isEqualToString:#"<sub2/>"])
{
replacement = #"₂";
}
If you use the OSX character viewer, you can drop unicode characters right into your code. There's a set of characters in there called "Digits" which has all the superscript and subscript number characters. Just leave your cursor at the appropriate spot in your code window and double-click in the character viewer to insert the character you want.
With the right font, you could probably do this with any letter as well, but the character map only has a handful of non-numbers available for this that I've seen.
Alternatively you can just put the unicode characters in your source content, but in a lot of cases (like mine), that isn't possible.
Swift 4
Very loosely based off of Graham Perks' answer. I could not make his code work as is but after three hours of work I've created something that works great! If you'd prefer a full implementation of this along with a bunch of nifty other performance and feature add-ons (links, async drawing, etc), check out my single file library DYLabel. If not, read on.
I explain everything I'm doing in the comments. This is the draw method, to be called from drawRect:
/// Draw text on a given context. Supports superscript using NSBaselineOffsetAttributeName
///
/// This method works by drawing the text backwards (i.e. last line first). This is very very important because it's how we ensure superscripts don't overlap the text above it. In other words, we need to start from the bottom, get the height of the text we just drew, and then draw the next text above it. This could be done in a forward direction but you'd have to use lookahead which IMO is more work.
///
/// If you have to modify on this, remember that CT uses a mathmatical origin (i.e. 0,0 is bottom left like a cartisian plane)
/// - Parameters:
/// - context: A core graphics draw context
/// - attributedText: An attributed string
func drawText(context:CGContext, attributedText: NSAttributedString) {
//Create our CT boiler plate
let framesetter = CTFramesetterCreateWithAttributedString(attributedText)
let textRect = bounds
let path = CGPath(rect: textRect, transform: nil)
let frame = CTFramesetterCreateFrame(framesetter, CFRangeMake(0, 0), path, nil)
//Fetch our lines, bridging to swift from CFArray
let lines = CTFrameGetLines(frame) as [AnyObject]
let lineCount = lines.count
//Get the line origin coordinates. These are used for calculating stock line height (w/o baseline modifications)
var lineOrigins = [CGPoint](repeating: CGPoint.zero, count: lineCount)
CTFrameGetLineOrigins(frame, CFRangeMake(0, 0), &lineOrigins);
//Since we're starting from the bottom of the container we need get our bottom offset/padding (so text isn't slammed to the bottom or cut off)
var ascent:CGFloat = 0
var descent:CGFloat = 0
var leading:CGFloat = 0
if lineCount > 0 {
CTLineGetTypographicBounds(lines.last as! CTLine, &ascent, &descent, &leading)
}
//This variable holds the current draw position, relative to CT origin of the bottom left
//https://stackoverflow.com/a/27631737/1166266
var drawYPositionFromOrigin:CGFloat = descent
//Again, draw the lines in reverse so we don't need look ahead
for lineIndex in (0..<lineCount).reversed() {
//Calculate the current line height so we can accurately move the position up later
let lastLinePosition = lineIndex > 0 ? lineOrigins[lineIndex - 1].y: textRect.height
let currentLineHeight = lastLinePosition - lineOrigins[lineIndex].y
//Throughout the loop below this variable will be updated to the tallest value for the current line
var maxLineHeight:CGFloat = currentLineHeight
//Grab the current run glyph. This is used for attributed string interop
let glyphRuns = CTLineGetGlyphRuns(lines[lineIndex] as! CTLine) as [AnyObject]
for run in glyphRuns {
let run = run as! CTRun
//Convert the format range to something we can match to our string
let runRange = CTRunGetStringRange(run)
let attribuetsAtPosition = attributedText.attributes(at: runRange.location, effectiveRange: nil)
var baselineAdjustment: CGFloat = 0.0
if let adjust = attribuetsAtPosition[NSAttributedStringKey.baselineOffset] as? NSNumber {
//We have a baseline offset!
baselineAdjustment = CGFloat(adjust.floatValue)
}
//Check if this glyph run is tallest, and move it if it is
maxLineHeight = max(currentLineHeight + baselineAdjustment, maxLineHeight)
//Move the draw head. Note that we're drawing from the unupdated drawYPositionFromOrigin. This is again thanks to CT cartisian plane where we draw from the bottom left of text too.
context.textPosition = CGPoint.init(x: lineOrigins[lineIndex].x, y: drawYPositionFromOrigin)
//Draw!
CTRunDraw(run, context, CFRangeMake(0, 0))
}
//Move our position because we've completed the drawing of the line which is at most `maxLineHeight`
drawYPositionFromOrigin += maxLineHeight
}
}
I also made a method which calculates the required height of the text given a width. It's exactly the same code except it doesn't draw anything.
/// Calculate the height if it were drawn using `drawText`
/// Uses the same code as drawText except it doesn't draw.
///
/// - Parameters:
/// - attributedText: The text to calculate the height of
/// - width: The constraining width
/// - estimationHeight: Optional paramater, default 30,000px. This is the container height used to layout the text. DO NOT USE CGFLOATMAX AS IT CORE TEXT CANNOT CREATE A FRAME OF THAT SIZE.
/// - Returns: The size required to fit the text
static func size(of attributedText:NSAttributedString,width:CGFloat, estimationHeight:CGFloat?=30000) -> CGSize {
let framesetter = CTFramesetterCreateWithAttributedString(attributedText)
let textRect = CGRect.init(x: 0, y: 0, width: width, height: estimationHeight!)
let path = CGPath(rect: textRect, transform: nil)
let frame = CTFramesetterCreateFrame(framesetter, CFRangeMake(0, 0), path, nil)
//Fetch our lines, bridging to swift from CFArray
let lines = CTFrameGetLines(frame) as [AnyObject]
let lineCount = lines.count
//Get the line origin coordinates. These are used for calculating stock line height (w/o baseline modifications)
var lineOrigins = [CGPoint](repeating: CGPoint.zero, count: lineCount)
CTFrameGetLineOrigins(frame, CFRangeMake(0, 0), &lineOrigins);
//Since we're starting from the bottom of the container we need get our bottom offset/padding (so text isn't slammed to the bottom or cut off)
var ascent:CGFloat = 0
var descent:CGFloat = 0
var leading:CGFloat = 0
if lineCount > 0 {
CTLineGetTypographicBounds(lines.last as! CTLine, &ascent, &descent, &leading)
}
//This variable holds the current draw position, relative to CT origin of the bottom left
var drawYPositionFromOrigin:CGFloat = descent
//Again, draw the lines in reverse so we don't need look ahead
for lineIndex in (0..<lineCount).reversed() {
//Calculate the current line height so we can accurately move the position up later
let lastLinePosition = lineIndex > 0 ? lineOrigins[lineIndex - 1].y: textRect.height
let currentLineHeight = lastLinePosition - lineOrigins[lineIndex].y
//Throughout the loop below this variable will be updated to the tallest value for the current line
var maxLineHeight:CGFloat = currentLineHeight
//Grab the current run glyph. This is used for attributed string interop
let glyphRuns = CTLineGetGlyphRuns(lines[lineIndex] as! CTLine) as [AnyObject]
for run in glyphRuns {
let run = run as! CTRun
//Convert the format range to something we can match to our string
let runRange = CTRunGetStringRange(run)
let attribuetsAtPosition = attributedText.attributes(at: runRange.location, effectiveRange: nil)
var baselineAdjustment: CGFloat = 0.0
if let adjust = attribuetsAtPosition[NSAttributedStringKey.baselineOffset] as? NSNumber {
//We have a baseline offset!
baselineAdjustment = CGFloat(adjust.floatValue)
}
//Check if this glyph run is tallest, and move it if it is
maxLineHeight = max(currentLineHeight + baselineAdjustment, maxLineHeight)
//Skip drawing since this is a height calculation
}
//Move our position because we've completed the drawing of the line which is at most `maxLineHeight`
drawYPositionFromOrigin += maxLineHeight
}
return CGSize.init(width: width, height: drawYPositionFromOrigin)
}
Like everything I write, I also did some benchmarks against some public libraries and system functions (even though they won't work here). I used a huge, complex string here to keep anyone from taking unfair shortcuts.
---HEIGHT CALCULATION---
Runtime for 1000 iterations (ms) BoundsForRect: 5415.030002593994
Runtime for 1000 iterations (ms) layoutManager: 5370.990991592407
Runtime for 1000 iterations (ms) CTFramesetterSuggestFrameSizeWithConstraints: 2372.151017189026
Runtime for 1000 iterations (ms) CTFramesetterCreateFrame ObjC: 2300.302028656006
Runtime for 1000 iterations (ms) CTFramesetterCreateFrame-Swift: 2313.6669397354126
Runtime for 1000 iterations (ms) THIS ANSWER size(of:): 2566.351056098938
---RENDER---
Runtime for 1000 iterations (ms) AttributedLabel: 35.032033920288086
Runtime for 1000 iterations (ms) UILabel: 45.948028564453125
Runtime for 1000 iterations (ms) TTTAttributedLabel: 301.1329174041748
Runtime for 1000 iterations (ms) THIS ANSWER: 20.398974418640137
So summary time: we did very well! size(of...) is nearly equal to stock CT layout which means that our addon for superscript is fairly cheap despite using a hash table lookup. We do, however, flat out win on draw calls. I suspect that this is due to the very expensive 30k pixel estimation frame we have to create. If we make a better estimate performance will be better. I've already been working for about three hours so I'm calling it quits and leaving that as an exercise to the reader.
I struggled with this problem as well. It turns out, as some of the posters above suggested, that none of the fonts that come with IOS support superscripting or subscripting. My solution was to purchase and install two custom superscript and subscript fonts (They were $9.99 each and here's a link to the site http://superscriptfont.com/).
Not really that hard to do. Just add the font files as resources and add info.plist entries for "Font provided by application".
The next step was to search for the appropriate tags in my NSAttributedString, remove the tags and apply the font to the text.
Works great!
A Swift 2 twist on Dimitri's answer; effectively implements NSBaselineOffsetAttributeName.
When coding I was in a UIView so had a reasonable bounds rect to use. His answer calculated its own rect.
func drawText(context context:CGContextRef, attributedText: NSAttributedString) {
// All this CoreText iteration just to add support for superscripting.
// NSBaselineOffsetAttributeName isn't supported by CoreText. So we manully iterate through
// all the text ranges, rendering each, and offsetting the baseline where needed.
let framesetter = CTFramesetterCreateWithAttributedString(attributedText)
let textRect = CGRectOffset(bounds, 0, 0)
let path = CGPathCreateWithRect(textRect, nil)
let frame = CTFramesetterCreateFrame(framesetter, CFRangeMake(0, 0), path, nil)
// All the lines of text we'll render...
let lines = CTFrameGetLines(frame) as [AnyObject]
let lineCount = lines.count
// And their origin coordinates...
var lineOrigins = [CGPoint](count: lineCount, repeatedValue: CGPointZero)
CTFrameGetLineOrigins(frame, CFRangeMake(0, 0), &lineOrigins);
for lineIndex in 0..<lineCount {
let lineObject = lines[lineIndex]
// Each run of glyphs we'll render...
let glyphRuns = CTLineGetGlyphRuns(lineObject as! CTLine) as [AnyObject]
for r in glyphRuns {
let run = r as! CTRun
let runRange = CTRunGetStringRange(run)
// What attributes are in the NSAttributedString here? If we find NSBaselineOffsetAttributeName,
// adjust the baseline.
let attrs = attributedText.attributesAtIndex(runRange.location, effectiveRange: nil)
var baselineAdjustment: CGFloat = 0.0
if let adjust = attrs[NSBaselineOffsetAttributeName as String] as? NSNumber {
baselineAdjustment = CGFloat(adjust.floatValue)
}
CGContextSetTextPosition(context, lineOrigins[lineIndex].x, lineOrigins[lineIndex].y - 25 + baselineAdjustment)
CTRunDraw(run, context, CFRangeMake(0, 0))
}
}
}
With IOS 11, Apple introduced a new string attribute name:
kCTBaselineOffsetAttributeName which works with Core Text.
Note that the offset direction is different from NSBaselineOffsetAttributeName used with NSAttributedStrings on UILabels etc (a positive offset moves the baseline downwards).