Skip to content

Handle definition lists. #128

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mwhapples merged 1 commit into from
Jun 3, 2026
Merged

Handle definition lists. #128

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
40 changes: 33 additions & 7 deletions brailleblaster-ebraille/src/main/kotlin/org/brailleblaster/ebraille/bbx2html/blocks.kt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,19 +21,20 @@ import org.brailleblaster.ebraille.asciiToEbraille
import org.brailleblaster.utils.xml.UTD_NS
import org.brailleblaster.utils.xom.childNodes

internal fun Element.processBlock(): Collection<org.jsoup.nodes.Element> = when(BBX.BLOCK.getSubType(this)) {
internal fun Element.processBlock(): Collection<org.jsoup.nodes.Element> = when (BBX.BLOCK.getSubType(this)) {
BBX.BLOCK.STYLE -> processStyle()
BBX.BLOCK.LIST_ITEM -> listOf(processParagraph(tag = "li"))
BBX.BLOCK.PAGE_NUM -> listOf(processPageNum())
BBX.BLOCK.DEFAULT -> listOf(processParagraph())
else -> listOf(processParagraph())
}

internal fun Element.processPageNum(): org.jsoup.nodes.Element = org.jsoup.nodes.Element("span").attr("role", "doc-pagebreak").apply {
val brl = getFirstChildElement("brl", UTD_NS)
attr("aria-label", brl.getAttributeValue("printPage").orEmpty().ifEmpty { "-" })
appendText(asciiToEbraille(brl.getAttributeValue("printPageBrl").orEmpty().ifEmpty { "\u2824" }))
}
internal fun Element.processPageNum(): org.jsoup.nodes.Element =
org.jsoup.nodes.Element("span").attr("role", "doc-pagebreak").apply {
val brl = getFirstChildElement("brl", UTD_NS)
attr("aria-label", brl.getAttributeValue("printPage").orEmpty().ifEmpty { "-" })
appendText(asciiToEbraille(brl.getAttributeValue("printPageBrl").orEmpty().ifEmpty { "\u2824" }))
}

private fun Element.processStyle(): Collection<org.jsoup.nodes.Element> = when (style) {
"Centered Heading" -> listOf(processParagraph(tag = "h1"))
Expand All @@ -51,4 +52,29 @@ private fun Element.processParagraph(
for ((k, v) in attributes) {
attr(k, v)
}
}.appendChildren(childNodes.flatMap { it.processContent() })
}.appendChildren(childNodes.flatMap { it.processContent() })

private sealed interface DefinitionListItem {
data class Term(val element: Element) : DefinitionListItem
data class Definition(val elements: List<Element>) : DefinitionListItem
}

internal fun Element.processDefinitionListItem(): List<org.jsoup.nodes.Element> =
childElements.fold(listOf<DefinitionListItem>()) { acc, element ->
if (BBX.SPAN.DEFINITION_TERM.isA(element)) {
acc + DefinitionListItem.Term(element)
} else {
val prev = acc.lastOrNull()
if (prev is DefinitionListItem.Definition) {
acc.dropLast(1) + DefinitionListItem.Definition(prev.elements + element)
} else {
acc + DefinitionListItem.Definition(listOf(element))
}
}
}.map {
when (it) {
is DefinitionListItem.Term -> it.element.processParagraph(tag = "dt")
is DefinitionListItem.Definition -> org.jsoup.nodes.Element("dd")
.appendChildren(it.elements.flatMap { e -> e.processContent() })
}
}
97 changes: 61 additions & 36 deletions brailleblaster-ebraille/src/main/kotlin/org/brailleblaster/ebraille/bbx2html/containers.kt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,51 +42,73 @@ private fun Element.processBox(): org.jsoup.nodes.Element {

private fun Element.processList(): org.jsoup.nodes.Element =
childElements.filter { BBX.BLOCK.LIST_ITEM.isA(it) }
.map { ListItem(it, it.getAttributeValue("itemLevel", BB_NS)?.toIntOrNull() ?: 0) }.toHtml(
level = 0,
containerFactory = { org.jsoup.nodes.Element("ul").attr("style", "list-style-type: none") }
) { it.element.processBlock() }

private fun Element.processTable(): List<org.jsoup.nodes.Element> = if (getAttributeValue("tableCopy", UTD_NS) == "true") {
listOf()
} else {
val tableFormat = getAttributeValue("format")
listOf(org.jsoup.nodes.Element("table").also {
if (tableFormat in listOf("listed", "stairstep", "linear")) {
it.attr("class", tableFormat)
}
}.appendChildren(
childElements.filter { BBX.CONTAINER.TABLE_ROW.isA(it) }.take(1).let { firstRow ->
when (tableFormat) {
"simple" -> {
firstRow.processTableRows(if (getAttributeValue("columnHeading") == "false") "td" else "th")
}
"listed" -> {
firstRow.processTableRows("th")
}
"stairstep" -> {
headerRowFromStairStepTableTN(this) + (firstRow.processTableRows())
.map { ListItem(it, it.getAttributeValue("itemLevel", BB_NS)?.toIntOrNull() ?: 0) }.let { items ->
when (BBX.CONTAINER.LIST.ATTRIB_LIST_TYPE.get(this)) {
BBX.ListType.DEFINITION -> items.toHtml(
level = 0,
containerFactory = { org.jsoup.nodes.Element("dl") }
) {
it.element.processDefinitionListItem()
}
"linear" -> {
headerRowFromLinearTableTN(this) + (firstRow.processTableRows())
}
else -> {
firstRow.processTableRows()
else -> items.toHtml(
level = 0,
containerFactory = { org.jsoup.nodes.Element("ul").attr("style", "list-style-type: none") }
) {
it.element.processBlock()
}
}
} + (childElements.filter { BBX.CONTAINER.TABLE_ROW.isA(it) }.drop(1).processTableRows())
))
}
}

private fun Element.processTable(): List<org.jsoup.nodes.Element> =
if (getAttributeValue("tableCopy", UTD_NS) == "true") {
listOf()
} else {
val tableFormat = getAttributeValue("format")
listOf(
org.jsoup.nodes.Element("table").also {
if (tableFormat in listOf("listed", "stairstep", "linear")) {
it.attr("class", tableFormat)
}
}.appendChildren(
childElements.filter { BBX.CONTAINER.TABLE_ROW.isA(it) }.take(1).let { firstRow ->
when (tableFormat) {
"simple" -> {
firstRow.processTableRows(if (getAttributeValue("columnHeading") == "false") "td" else "th")
}

"listed" -> {
firstRow.processTableRows("th")
}

"stairstep" -> {
headerRowFromStairStepTableTN(this) + (firstRow.processTableRows())
}

"linear" -> {
headerRowFromLinearTableTN(this) + (firstRow.processTableRows())
}

private fun Iterable<Element>.processTableRows(cellTag: String = "td"): List<org.jsoup.nodes.Element> = map { r -> org.jsoup.nodes.Element("tr").appendChildren(r.childElements.filter { BBX.BLOCK.TABLE_CELL.isA(it) }.map { c -> org.jsoup.nodes.Element(cellTag).appendChildren(c.processContent())}) }
else -> {
firstRow.processTableRows()
}
}
} + (childElements.filter { BBX.CONTAINER.TABLE_ROW.isA(it) }.drop(1).processTableRows())
)
)
}

private fun Iterable<Element>.processTableRows(cellTag: String = "td"): List<org.jsoup.nodes.Element> = map { r ->
org.jsoup.nodes.Element("tr").appendChildren(r.childElements.filter { BBX.BLOCK.TABLE_CELL.isA(it) }
.map { c -> org.jsoup.nodes.Element(cellTag).appendChildren(c.processContent()) })
}

private fun headerRowFromStairStepTableTN(table: Element): List<org.jsoup.nodes.Element> =
(table.previousSibling { it is Element } as? Element)?.let { e ->
if (BBX.CONTAINER.TABLETN.isA(e)) {
listOf(
org.jsoup.nodes.Element("tr").appendChildren(
e.childElements.filter { BBX.BLOCK.isA(it) }.drop(1)
.map { org.jsoup.nodes.Element("th").appendChildren(it.processContent()) })
.map { org.jsoup.nodes.Element("th").appendChildren(it.processContent()) })
)
} else {
listOf()
Expand All @@ -98,8 +120,11 @@ private fun headerRowFromLinearTableTN(table: Element): List<org.jsoup.nodes.Ele
if (BBX.CONTAINER.TABLETN.isA(e)) {
listOf(
org.jsoup.nodes.Element("tr").appendChildren(
e.childElements.filter { BBX.BLOCK.isA(it) }.takeLast(1).flatMap { b -> b.childElements.filter { BBX.SPAN.OTHER.isA(it) }.map { org.jsoup.nodes.Element("th").appendChildren(it.processContent()) } }
)
e.childElements.filter { BBX.BLOCK.isA(it) }.takeLast(1).flatMap { b ->
b.childElements.filter { BBX.SPAN.OTHER.isA(it) }
.map { org.jsoup.nodes.Element("th").appendChildren(it.processContent()) }
}
)
)
} else {
listOf()
Expand Down
Loading