changeset
Add lxp.totable by Tom?s Guisasola
Fri, 23 Apr 2021 21:18:24 +0100
- author
- Matthew Wild <mwild1@gmail.com>
- date
- Fri, 23 Apr 2021 21:18:24 +0100
- changeset 37
- 233463804681
- parent 36
- 4a61f00ee916
- child 38
- 1ad8000fedc0
Add lxp.totable by Tom?s Guisasola
| src/lxp/totable.lua | file | annotate | diff | comparison | revisions | |
| tests/test-table.lua | file | annotate | diff | comparison | revisions |
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/src/lxp/totable.lua Fri Apr 23 21:18:24 2021 +0100 @@ -0,0 +1,117 @@ +-- See Copyright Notice in license.html +-- Based on Luiz Henrique de Figueiredo's lxml: +-- http://www.tecgraf.puc-rio.br/~lhf/ftp/lua/#lxml + +local lxp = require "lxp" + +local table = require"table" +local tinsert, tremove = table.insert, table.remove +local assert, pairs, tostring, type = assert, pairs, tostring, type + +-- auxiliary functions ------------------------------------------------------- +local function starttag (p, tag, attr) + local stack = p:getcallbacks().stack + local newelement = {[0] = tag} + for i = 1, #attr do + local attrname = attr[i] + local attrvalue = attr[attrname] + newelement[attrname] = attrvalue + end + tinsert(stack, newelement) +end + +local function endtag (p, tag) + local stack = p:getcallbacks().stack + local element = tremove(stack) + assert(element[0] == tag, "Error while closing element: table[0] should be `"..tostring(tag).."' but is `"..tostring(element[0]).."'") + local level = #stack + tinsert(stack[level], element) +end + +local function text (p, txt) + local stack = p:getcallbacks().stack + local element = stack[#stack] + local n = #element + if type(element[n]) == "string" and n > 0 then + element[n] = element[n] .. txt + else + tinsert(element, txt) + end +end + +-- main function ------------------------------------------------------------- +local function parse (o) + local c = { + StartElement = starttag, + EndElement = endtag, + CharacterData = text, + _nonstrict = true, + stack = {{}}, + } + local p = lxp.new(c) + if type(o) == "string" then + local status, err, line, col, pos = p:parse(o) + if not status then return nil, err, line, col, pos end + else + for l in pairs(o) do + local status, err, line, col, pos = p:parse(l) + if not status then return nil, err, line, col, pos end + end + end + local status, err, line, col, pos = p:parse() -- close document + if not status then return nil, err, line, col, pos end + p:close() + return c.stack[1][1] +end + +-- utility functions --------------------------------------------------------- +local function compact (t) -- remove empty entries + local n = 0 + for i = 1, #t do + local v = t[i] + if v then + n = n+1 + if n ~= i then + t[n] = v + t[i] = nil + end + else + t[i] = nil + end + end +end + +local function clean (t) -- remove empty strings + for i = 1, #t do + local v = t[i] + local tv = type(v) + if tv == "table" then + clean (v) + elseif tv == "string" and v:match"^%s*$" then + t[i] = false + end + end + compact (t) +end + +local function torecord (t) -- move 1-value subtables to table entries + for i = 1, #t do + local v = t[i] + if type(v) == "table" then + if #v == 1 and type(v[1]) == "string" and t[v[0]] == nil then + t[v[0]] = v[1] + t[i] = false + else + torecord (v) + end + end + end + compact (t) +end + +return { + clean = clean, + compact = compact, + parse = parse, + torecord = torecord, +}
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/test-table.lua Fri Apr 23 21:18:24 2021 +0100 @@ -0,0 +1,278 @@ +#!/usr/local/bin/lua + +local totable = require "lxp.totable" + +local tests = { + { + [[<abc a1="A1" a2="A2">inside tag `abc'</abc>]], + { + [0] = "abc", + a1 = "A1", + a2 = "A2", + "inside tag `abc'", + }, + }, + { + [[<qwerty q1="q1" q2="q2"> + <asdf>some text</asdf> +</qwerty>]], + { + [0] = "qwerty", + q1 = "q1", + q2 = "q2", + "\n\t", + { + [0] = "asdf", + "some text", + }, + "\n", + }, + }, + { + [[ +<!-- http://www.w3schools.com/xml/simple.xml --> +<breakfast_menu> + <food> + <name>Belgian Waffles</name> + <price>$5.95</price> + <description>Two of our famous Belgian Waffles with plenty of real maple syrup</description> + <calories>650</calories> + </food> + <food> + <name>Strawberry Belgian Waffles</name> + <price>$7.95</price> + <description>Light Belgian waffles covered with strawberries and whipped cream</description> + <calories>900</calories> + </food> + <food> + <name>Berry-Berry Belgian Waffles</name> + <price>$8.95</price> + <description>Light Belgian waffles covered with an assortment of fresh berries and whipped cream</description> + <calories>900</calories> + </food> + <food> + <name>French Toast</name> + <price>$4.50</price> + <description>Thick slices made from our homemade sourdough bread</description> + <calories>600</calories> + </food> + <food> + <name>Homestyle Breakfast</name> + <price>$6.95</price> + <description>Two eggs, bacon or sausage, toast, and our ever-popular hash browns</description> + <calories>950</calories> + </food> +</breakfast_menu>]], + { + [0] = "breakfast_menu", + [1] = "\n\t", + [2] = { + [0] = "food", + [1] = "\n\t\t", + [2] = { [0] = "name", [1] = "Belgian Waffles", }, + [3] = "\n\t\t", + [4] = { [0] = "price", [1] = "$5.95", }, + [5] = "\n\t\t", + [6] = { + [0] = "description", + [1] = "Two of our famous Belgian Waffles with plenty of real maple syrup", + }, + [7] = "\n\t\t", + [8] = { [0] = "calories", [1] = "650", }, + [9] = "\n\t", + }, + [3] = "\n\t", + [4] = { + [0] = "food", + [1] = "\n\t\t", + [2] = { [0] = "name", [1] = "Strawberry Belgian Waffles", }, + [3] = "\n\t\t", + [4] = { [0] = "price", [1] = "$7.95", }, + [5] = "\n\t\t", + [6] = { + [0] = "description", + [1] = "Light Belgian waffles covered with strawberries and whipped cream", + }, + [7] = "\n\t\t", + [8] = { [0] = "calories", [1] = "900", }, + [9] = "\n\t", + }, + [5] = "\n\t", + [6] = { + [0] = "food", + [1] = "\n\t\t", + [2] = { [0] = "name", [1] = "Berry-Berry Belgian Waffles", }, + [3] = "\n\t\t", + [4] = { [0] = "price", [1] = "$8.95", }, + [5] = "\n\t\t", + [6] = { + [0] = "description", + [1] = "Light Belgian waffles covered with an assortment of fresh berries and whipped cream", + }, + [7] = "\n\t\t", + [8] = { [0] = "calories", [1] = "900", }, + [9] = "\n\t", + }, + [7] = "\n\t", + [8] = { + [0] = "food", + [1] = "\n\t\t", + [2] = { [0] = "name", [1] = "French Toast", }, + [3] = "\n\t\t", + [4] = { [0] = "price", [1] = "$4.50", }, + [5] = "\n\t\t", + [6] = { + [0] = "description", + [1] = "Thick slices made from our homemade sourdough bread", + }, + [7] = "\n\t\t", + [8] = { [0] = "calories", [1] = "600", }, + [9] = "\n\t", + }, + [9] = "\n\t", + [10] = { + [0] = "food", + [1] = "\n\t\t", + [2] = { [0] = "name", [1] = "Homestyle Breakfast", }, + [3] = "\n\t\t", + [4] = { [0] = "price", [1] = "$6.95", }, + [5] = "\n\t\t", + [6] = { + [0] = "description", + [1] = "Two eggs, bacon or sausage, toast, and our ever-popular hash browns", + }, + [7] = "\n\t\t", + [8] = { [0] = "calories", [1] = "950", }, + [9] = "\n\t", + }, + [11] = "\n", + }, + clean = { + [0] = "breakfast_menu", + [1] = { + [0] = "food", + [1] = { [0] = "name", [1] = "Belgian Waffles", }, + [2] = { [0] = "price", [1] = "$5.95", }, + [3] = { + [0] = "description", + [1] = "Two of our famous Belgian Waffles with plenty of real maple syrup", + }, + [4] = { [0] = "calories", [1] = "650", }, + }, + [2] = { + [0] = "food", + [1] = { [0] = "name", [1] = "Strawberry Belgian Waffles", }, + [2] = { [0] = "price", [1] = "$7.95", }, + [3] = { + [0] = "description", + [1] = "Light Belgian waffles covered with strawberries and whipped cream", + }, + [4] = { [0] = "calories", [1] = "900", }, + }, + [3] = { + [0] = "food", + [1] = { [0] = "name", [1] = "Berry-Berry Belgian Waffles", }, + [2] = { [0] = "price", [1] = "$8.95", }, + [3] = { + [0] = "description", + [1] = "Light Belgian waffles covered with an assortment of fresh berries and whipped cream", + }, + [4] = { [0] = "calories", [1] = "900", }, + }, + [4] = { + [0] = "food", + [1] = { [0] = "name", [1] = "French Toast", }, + [2] = { [0] = "price", [1] = "$4.50", }, + [3] = { + [0] = "description", + [1] = "Thick slices made from our homemade sourdough bread", + }, + [4] = { [0] = "calories", [1] = "600", }, + }, + [5] = { + [0] = "food", + [1] = { [0] = "name", [1] = "Homestyle Breakfast", }, + [2] = { [0] = "price", [1] = "$6.95", }, + [3] = { + [0] = "description", + [1] = "Two eggs, bacon or sausage, toast, and our ever-popular hash browns", + }, + [4] = { [0] = "calories", [1] = "950", }, + }, + }, + torecord = { + [0] = "breakfast_menu", + [1] = { + [0] = "food", + name = "Belgian Waffles", + price = "$5.95", + description = "Two of our famous Belgian Waffles with plenty of real maple syrup", + calories = "650", + }, + [2] = { + [0] = "food", + name = "Strawberry Belgian Waffles", + price = "$7.95", + description = "Light Belgian waffles covered with strawberries and whipped cream", + calories = "900", + }, + [3] = { + [0] = "food", + name = "Berry-Berry Belgian Waffles", + price = "$8.95", + description = "Light Belgian waffles covered with an assortment of fresh berries and whipped cream", + calories = "900", + }, + [4] = { + [0] = "food", + name = "French Toast", + price = "$4.50", + description = "Thick slices made from our homemade sourdough bread", + calories = "600", + }, + [5] = { + [0] = "food", + name = "Homestyle Breakfast", + price = "$6.95", + description = "Two eggs, bacon or sausage, toast, and our ever-popular hash browns", + calories = "950", + }, + }, + }, +} + + +function table.equal (t1, t2) + for nome, val in pairs (t1) do + local tv = type(val) + if tv == "table" then + if type(t2[nome]) ~= "table" then + return false, "Different types at entry `"..nome.."': t1."..nome.." is "..tv.." while t2."..nome.." is "..type(t2[nome]).." ["..tostring(t2[nome]).."]" + else + local ok, msg = table.equal (val, t2[nome]) + if not ok then + return false, "["..nome.."]\t"..tostring(val).." ~= "..tostring(t2[nome]).."; "..msg + end + end + else + if val ~= t2[nome] then + return false, "["..nome.."]\t"..tostring(val).." ~= "..tostring(t2[nome]) + end + end + end + return true +end + + +for i, s in ipairs(tests) do + local ds = assert (totable.parse ([[<?xml version="1.0" encoding="ISO-8859-1"?>]]..s[1])) + assert(table.equal (ds, s[2])) +end + +local t = totable.parse ([[<?xml version="1.0" encoding="ISO-8859-1"?>]]..tests[3][1]) +totable.clean (t) +assert (table.equal (t, tests[3].clean)) +totable.torecord (t) +assert (table.equal (t, tests[3].torecord)) + +print"OK"
