# -*- coding: utf-8 -*-
"""
Created on Thu Aug 3 21:13:51 2023

@author: Arnd Helmut Hafner
"""

#本スクリプトは、釈文と様式項目を別々のテキストファイルから読み込み、様式項目を見出し行に含めたHTMLファイルに作り替える。ヘッダーとフッターに必要な情報は別途head.txtとfoot.txtから読み取る。
import re
import sys # for sys.exit after error

#read shakumon-file
datafile = open('shakumon.txt', 'r', encoding='utf-8-sig')
yoshiki = [] #様式分類を保存（書式は「n-n-n-n-n-n」)
haichi = [] #様式分類内の配置位置・順番を記憶する
kanmen = [] #同一の文書が記されている簡面を数える
kanbango = [] #簡番号を記憶する
shakumon = [] #store text
recordcount = -1
records = {} #A dictionary of lists of records with the same yoshikibango; yoshikibango is used as key, the record number as value
recordkeys = [] #Gathering the keys of the record-dictionary. It actually comprises a list of yoshikibango. 
linecount = 0
preshakumon = True #signals preexisting line of shakumon
for line in datafile: #reads data line by line, but storing data record by record
    linecount += 1
    m = re.search('^(\d+\-\d+\-\d+\-\d+\-\d+\-\d+)\t(\d+)\t(\d+)\t([\d\-\+abcd=J⑦⑨⑩⑫⑭⑮⑯⑰K（）]+)$',line)
    if m: #detecting the first line of a record and gathering yoshiki etc
        recordcount += 1
        yoshiki.append(m.group(1))
        haichi.append(m.group(2))
        kanmen.append(m.group(3))
        kanbango.append(m.group(4))
        preshakumon = False
        if yoshiki[recordcount] in recordkeys:#adding the record number to the list of records if the record key exists 
            records[yoshiki[recordcount]].append(recordcount)
        else:#create a new key to the record list and add the first record number to the empty list
            recordkeys.append(yoshiki[recordcount])
            records[yoshiki[recordcount]] = [recordcount]
    elif re.search('^[^\d]', line):#detecting shakumon
        if preshakumon:#second line of shakumon and below
            shakumon[recordcount] = shakumon[recordcount] + line
        else:#first line of shakumon
            shakumon.append(line)
            preshakumon = True
    else:#testing mismatches
        sys.exit('mismatch in line'+str(linecount)+':'+line)
datafile.close()

#read 様式分類項目
yoshikifile = open('yoshikikomoku.txt', 'r', encoding='utf-8-sig')
yoshikibango = [] #様式分類記号を蓄積する
yoshikikomoku = [] #様式分類項目を蓄積する
linecount = 0
for line in yoshikifile: #reads data line by line
    linecount += 1
    m = re.search('^(\d+\-\d+\-\d+\-\d+\-\d+\-\d+)\t([^\t\n]+)$',line)
    if m:
        yoshikibango.append(m.group(1))
        yoshikikomoku.append(m.group(2))
    else:#testing mismatches
        sys.exit('mismatch in line'+str(linecount)+':'+line)
yoshikifile.close


#reading and writing the header
outfile = open('liya-shakumon(python).html', 'w', encoding='utf-8-sig')
headfile = open('head.txt', 'r', encoding='utf-8-sig')
for line in headfile: outfile.write(line)


#write navigation list
out ='' #出力用の加工データを記憶する。
head = '\n<header id="menu">\n<div id="Container1">\n<ul id="MenuBar1" class="MenuBarHorizontal">\n'#ナビゲーションメニューの最初の三行
outfile.write(head)
foot = '</div>\n</header>\n'
isoshin = 0 #現在の位相を記憶する
isokyu = 0 #一つ前の位相を記憶する
indento = ['\t','\t\t','\t\t\t','\t\t\t\t','\t\t\t\t\t','\t\t\t\t\t\t']#位相に合わせたインデントを作るために、タブ文字のリストを記憶する。
listcontainermae = '<ul>\n'
listcontainerato = '</ul></li>\n'
listkomokumae = '<li><a href="#'
listkomokunaka = '">'
listkomokuato = '</a></li>\n'
for i in range(len(yoshikibango)):
    isokyu = isoshin #位相比較のため前の様式項目の位相を記憶する
    isoshin = 6-len(re.findall('\-0',yoshikibango[i]))#新しい項目の位相を「-0」の検出頻度で計算する
    if isoshin == isokyu:#位相に変化がなければ、outに単純に新しいリスト項目を加える。
        out += indento[isoshin-1] + listkomokumae + yoshikibango[i] +listkomokunaka + yoshikikomoku[i] + listkomokuato
    elif isoshin > isokyu:#位相が上がるたび、最後のリスト項目から末尾の「</li>」を削除した上、新しいリストタグ「<ul>を加える
        out = re.sub('</li>\n$','\n' + indento[isoshin-2] + listcontainermae,out)
        out += indento[isoshin-1] + listkomokumae + yoshikibango[i] +listkomokunaka + yoshikikomoku[i] + listkomokuato
    elif isoshin < isokyu:#位相が下がるたび、前のリストを締めて、より位相の低いリスト項目を一つ加える。
        while isoshin < isokyu:#位相が二つ以上下がることもあるので、関連するリストを一つづつ「</ul></li>」で閉じる。
            out += indento[isokyu-2] + listcontainerato
            isokyu -= 1
        out += indento[isoshin-1] + listkomokumae + yoshikibango[i] +listkomokunaka + yoshikikomoku[i] + listkomokuato
out += indento[isokyu-2] + listcontainerato#最後のリストを閉じる
out += foot
outfile.write(out)


#write text data
outfile.write('\n<body>/n<p>　</p>\n')
h = 0 #標題行の位相を記憶する
refmae = '<p>　　　　<a href="Liya-shakudoku/'
refnaka = '.html" target="_blank">釈読情報</a>　　　　<a href="Liya-yomikudashi/'
refusiro = '.html" target="_blank">読下文と注釈</a></p>'
space ='<p>　</p>\n\n'
for i in range(len(yoshikibango)):
    h = 7-len(re.findall('\-0',yoshikibango[i]))
    out = '<h'+str(h)+' id="'+yoshikibango[i]+'">'+yoshikikomoku[i]+'</h'+str(h)+'>\n'
    outfile.write(out)
    if yoshikibango[i] in recordkeys:
        liste = records[yoshikibango[i]]
        liste.sort(reverse=True)
        text = ''
        for l in liste:
            honbun = re.sub('\n','</p>\n<p>',shakumon[l])
            honbun = re.sub('</p>\n<p>$','\t'+kanbango[l]+'</p>\n',honbun)
            honbun = '<p>'+honbun
            if (kanmen[l] == '0') and (l+1 not in liste):
                honbun = honbun+refmae+kanbango[l]+refnaka+kanbango[l]+refusiro+'\n'+space
            elif l+1 not in liste:
                bango = ''
                n = int(kanmen[l])
                while n > 0:
                    bango = bango + kanbango[l-n] + '='
                    n -=1
                bango = bango + kanbango[l]
                honbun = honbun+refmae+bango+refnaka+bango+refusiro+'\n'+space
            elif int(kanmen[l]) > int(kanmen[l+1]):
                bango = ''
                n = int(kanmen[l])
                while n > 0:
                    bango = bango + kanbango[l-n] + '='
                    n -=1
                bango = bango + kanbango[l]
                honbun = honbun+refmae+bango+refnaka+bango+refusiro+'\n'+space
            elif int(kanmen[l]) < int(kanmen[l+1]):
                honbun = honbun
            elif kanmen[l] == '0' and kanmen[l+1] == '0':
                honbun = honbun+refmae+kanbango[l]+refnaka+kanbango[l]+refusiro+'\n'+space
            else:
                print('mismatch for '+kanbango[l])
            text = honbun + text    
        outfile.write(text)

#write footer
footfile = open('foot.txt', 'r', encoding='utf-8-sig')
for line in footfile: outfile.write(line)



outfile.close()