statistics: consider overhang of horizontal axes

The old code didn't consider that labels can peak out of
horizontal axes if labels are under ticks.

This commit takes this into account. However, it must be
noted that this is only heuristics: Before setting the
size of the axes, the actual minimum and maximum label are
not known, because we round to "nice" numbers. But the
size of the axis can only be set after knowing the overhang,
leading to a circular dependency. Therefore, the code
currently simply uses the minimum and maximum value of
the data, hoping that the "nice" values will not format
to something significantly larger. We could do a multi-pass
scheme, but let's not for now.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>

1 files changed, 48 insertions, 8 deletions

diff --git a/stats/statsaxis.cpp b/stats/statsaxis.cpp
index 8b8694964..d31b5827b 100644
--- a/stats/statsaxis.cpp
+++ b/stats/statsaxis.cpp
@@ -57,6 +57,17 @@ std::pair<double, double> StatsAxis::minMaxScreen() const
 			  : std::make_pair(zeroOnScreen, zeroOnScreen - size);
 }
 
+std::pair<double, double> StatsAxis::horizontalOverhang() const
+{
+	// If the labels are between ticks, they cannot peak out
+	if (!horizontal || labelsBetweenTicks)
+		return { 0.0, 0.0 };
+	QFontMetrics fm(labelFont);
+	auto [firstLabel, lastLabel] = getFirstLastLabel();
+	return { fm.size(Qt::TextSingleLine, firstLabel).width() / 2.0,
+		 fm.size(Qt::TextSingleLine, lastLabel).width() / 2.0 };
+}
+
 void StatsAxis::setRange(double minIn, double maxIn)
 {
 	min = minIn;
@@ -223,6 +234,19 @@ ValueAxis::ValueAxis(const QString &title, double min, double max, int decimals,
 	StatsAxis(title, horizontal, false),
 	min(min), max(max), decimals(decimals)
 {
+	// Avoid degenerate cases
+	if (max - min < 0.0001) {
+		max += 0.5;
+		min -= 0.5;
+	}
+}
+
+// Attention: this is only heuristics. Before setting the actual size, we
+// don't know the actual numbers of the minimum and maximum value.
+std::pair<QString, QString> ValueAxis::getFirstLastLabel() const
+{
+	QLocale loc;
+	return { loc.toString(min, 'f', decimals), loc.toString(max, 'f', decimals) };
 }
 
 void ValueAxis::updateLabels()
@@ -230,15 +254,8 @@ void ValueAxis::updateLabels()
 	labels.clear();
 	ticks.clear();
 
-	// Avoid degenerate cases
-	if (max - min < 0.0001) {
-		max += 0.5;
-		min -= 0.5;
-	}
-
 	QLocale loc;
-	QString minString = loc.toString(min, 'f', decimals);
-	QString maxString = loc.toString(max, 'f', decimals);
+	auto [minString, maxString] = getFirstLastLabel();
 	int numTicks = guessNumTicks({ minString, maxString});
 
 	// Use full decimal increments
@@ -279,6 +296,14 @@ CountAxis::CountAxis(const QString &title, int count, bool horizontal) :
 {
 }
 
+// Attention: this is only heuristics. Before setting the actual size, we
+// don't know the actual numbers of the minimum and maximum value.
+std::pair<QString, QString> CountAxis::getFirstLastLabel() const
+{
+	QLocale loc;
+	return { QString("0"), loc.toString(count) };
+}
+
 void CountAxis::updateLabels()
 {
 	labels.clear();
@@ -333,6 +358,13 @@ CategoryAxis::CategoryAxis(const QString &title, const std::vector<QString> &lab
 	setRange(-0.5, static_cast<double>(labels.size()) + 0.5);
 }
 
+// No implementation because the labels are inside ticks and this
+// is only used to calculate the "overhang" of labels under ticks.
+std::pair<QString, QString> CategoryAxis::getFirstLastLabel() const
+{
+	return { QString(), QString() };
+}
+
 void CategoryAxis::updateLabels()
 {
 	// TODO: paint ellipses if space too small
@@ -370,6 +402,14 @@ HistogramAxis::HistogramAxis(const QString &title, std::vector<HistogramAxisEntr
 	setRange(bin_values.front().value, bin_values.back().value);
 }
 
+std::pair<QString, QString> HistogramAxis::getFirstLastLabel() const
+{
+	if (bin_values.empty())
+		return { QString(), QString() };
+	else
+		return { bin_values.front().name, bin_values.back().name };
+}
+
 // Initialize a histogram axis with the given labels. Labels are specified as (name, value, recommended) triplets.
 // If labels are skipped, try to skip it in such a way that a recommended label is shown.
 // The one example where this is relevant is the quarterly bins, which are formated as (2019, q1, q2, q3, 2020, ...).