Handle spaces or tabs as line-prefixes in .TextGrid files

NEWS.md

@@ -8,6 +8,9 @@
   encoding of file is guessed using `readr::guess_encoding`; however, this
   can be overridden by identifying the encoding explicitly (e.g., 
   `encoding = "UTF-8"` or `encoding "UTF-16BE"`).
+* Some versions of Praat write TextGrid files with spaces; other versions write
+  with tabs. Previous versions of `textgRid` assumed spaces, but now both 
+  formats can be handled.
 
 
 # v1.0.1.9001

R/IntervalTier-utilities.R

@@ -1,20 +1,20 @@
 
 # Extract the start times of the intervals within a block of @praatText.
-.IntervalStartTimes <- function(praatText, pattern = '^ {12}xmin') {
+.IntervalStartTimes <- function(praatText, pattern = '^( {12}|\t{3})xmin') {
   .start_times <- .TierTimes(praatText, pattern)
   return(.start_times)
 }
 
 
 # Extract the end times of the intervals within a block of @praatText.
-.IntervalEndTimes <- function(praatText, pattern = '^ {12}xmax') {
+.IntervalEndTimes <- function(praatText, pattern = '^( {12}|\t{3})xmax') {
   .end_times <- .TierTimes(praatText, pattern)
   return(.end_times)
 }
 
 
 # Extract the text-mark values (interval labels) from a block of @praatText.
-.IntervalLabels <- function(praatText, pattern = '^ {12}(text|mark)') {
+.IntervalLabels <- function(praatText, pattern = '^( {12}|\t{3})(text|mark)') {
   .labels <- .TierLabels(praatText, pattern)
   return(.labels)
 }

R/PointTier-utilities.R

@@ -1,13 +1,13 @@
 
 # Extract the times of the points within a block of @praatText.
-.PointTimes <- function(praatText, pattern = '^ {12}(time|number)') {
+.PointTimes <- function(praatText, pattern = '^( {12}|\t{3})(time|number)') {
   .times <- .TierTimes(praatText, pattern)
   return(.times)
 }
 
 
 # Extract the labels of the points within a block of @praatText.
-.PointLabels <- function(praatText, pattern = '^ {12}(text|mark)') {
+.PointLabels <- function(praatText, pattern = '^( {12}|\t{3})(text|mark)') {
   .labels <- .TierLabels(praatText, pattern)
   return(.labels)
 }

R/TextGrid-constructor.R

@@ -49,7 +49,7 @@ setMethod(
       .textgrid <- textGrid
     }
     new(Class = 'TextGrid',
-        .PraatText2TierObjects(.textgrid),
+        .PraatText2TierObjects(.textgrid, pattern = "^( {4}|\t)item"),
         startTime = .TextGridTime(.textgrid, pattern = '^xmin'),
         endTime = .TextGridTime(.textgrid, pattern = '^xmax')
     )

R/TextGrid-utilities.R

@@ -24,7 +24,7 @@ NULL
 
 # Partition a block of @praatText into a list of sub-blocks, each sub-block
 # corresponding to the lines that define a tier within @praatText.
-.SplitPraatTextIntoTiers <- function(praatText, pattern = '^ {4}item') {
+.SplitPraatTextIntoTiers <- function(praatText, pattern = '^( {4}|\t)item') {
   .tier_blocks <- Map(
     `[`,
     rep(list(praatText), .CountTiers(praatText, pattern)),
@@ -49,7 +49,7 @@ NULL
 
 
 # The real engine for creating TextGrid objects.
-.PraatText2TierObjects <- function(praatText, pattern = '^ {4}item') {
+.PraatText2TierObjects <- function(praatText, pattern = '^( {4}|\t)item') {
   .tier_text    <- .SplitPraatTextIntoTiers(praatText, pattern)
   .tier_classes <- .TierClass(praatText)
   .tier_objects <- Map(.TierText2TierObject, .tier_text, .tier_classes)

R/Tier-utilities.R

@@ -8,7 +8,7 @@
 
 
 # Count the number of tiers that occur within a block of @praatText.
-.CountTiers <- function(praatText, pattern = '^ {4}item') {
+.CountTiers <- function(praatText, pattern = '^( {4}|\t)item') {
   .count <- length(.TierStartLine(praatText, pattern))
   return(.count)
 }
@@ -22,7 +22,7 @@
 
 
 # Extract the classes of the tiers that occur within a block of @praatText.
-.TierClass <- function(praatText, pattern = '^ {8}class') {
+.TierClass <- function(praatText, pattern = '^( {8}|\t{2})class') {
   .tier_classes <- .Extract(
     .Extract(.PraatLines(praatText, pattern),
                     pattern = '".*"'),
@@ -34,7 +34,7 @@
 
 # Find the line numbers that mark the ends of tiers within a block
 # of @praatText.
-.TierEndLine <- function(praatText, pattern = '^ {4}item') {
+.TierEndLine <- function(praatText, pattern = '^( {4}|\t)item') {
   .end_lines <- `[`(
     c(.TierStartLine(praatText, pattern) - 1, length(praatText)),
     2:(.CountTiers(praatText, pattern) + 1)
@@ -45,7 +45,7 @@
 
 # Generate a list of index-vectors. Each index-vector denotes the lines
 # within a block of @praatText that are spanned by a tier.
-.TierIndices <- function(praatText, pattern = '^ {4}item') {
+.TierIndices <- function(praatText, pattern = '^( {4}|\t)item') {
   .tier_indices <- Map(
     `:`,
     .TierStartLine(praatText, pattern),
@@ -56,7 +56,7 @@
 
 
 # Extract the names of the tiers that occur within a block of @praatText.
-.TierName <- function(praatText, pattern = '^ {8}name') {
+.TierName <- function(praatText, pattern = '^( {8}|\t{2})name') {
   .tier_names <- .Extract(
     .Extract(.PraatLines(praatText, pattern),
                     pattern = '".*"'),
@@ -67,7 +67,7 @@
 
 
 # Extract the numbers of the tiers that occur within a block of @praatText.
-.TierNumber <- function(praatText, pattern = '^ {4}item') {
+.TierNumber <- function(praatText, pattern = '^( {4}|\t)item') {
   .tier_numbers <- .Extract(
     .Extract(.PraatLines(praatText, pattern),
                     pattern = '\\[.*\\]'),
@@ -80,7 +80,7 @@
 
 # Find the line numbers that mark the beginnings of tiers within
 # a block of @praatText.
-.TierStartLine <- function(praatText, pattern = '^ {4}item') {
+.TierStartLine <- function(praatText, pattern = '^( {4}|\t)item') {
   .start_lines <- grep(pattern = pattern, x = praatText)
   return(.start_lines)
 }