[UPD] Cinemagia.ro

TekmanRO · Post by **TekmanRO** » 2005-06-16 12:33:46

Hi, I received a few emails from ppl asking me to update the CineMagia.RO script for 3.5.0, and make it retrieve some additional fields. So, I had some time to spare for it today, and decided to give it a go. So here's the updated CineMagia.RO script:

Code: Select all

(***************************************************

Ant Movie Catalog importation script
www.antp.be/software/moviecatalog/

[Infos]
Authors=Mihai Tudor (<link>TekmanRO@gmail.com</link>)
Title=Cinemagia.ro
Description=Cinemagia.ro (RO) import with picture
Site=http://www.cinemagia.ro
Language=RO
Version=0.9
Requires=3.5.0
Comments=The script imports the first article (usually Sinopsis) into the description field, and all the other articles into the Comments section.|Also, some xHTML formatting tags will appear in the text. They might not look nice in the AMC GUI, but once exported back to HTML, its pretty neat :)
License=This program is free software; you can redistribute it and/or modify it under the  terms of the GNU General Public License as published by the Free Software Foundation;  either version 2 of the License, or (at your option) any later version. |
GetInfo=1

[Options]

***************************************************)

program Cinemagia;

var
  MovieName: string;
  MovieURL: string;

function FindLine(Pattern: string; List: TStringList; StartAt: Integer): Integer;
var
  i: Integer;
begin
  Result := -1;
  if StartAt < 0 then
    StartAt := 0;
  for i := StartAt to List.Count-1 do
    if Pos(Pattern, List.GetString(i)) <> 0 then
    begin
      Result := i;
      Break;
    end;
end;

procedure AnalyzePage(Address: string);
var
  Page: TStringList;
begin
  Page := TStringList.Create;
  Page.Text := GetPage(Address);

  if pos('Rezultatele cautarii', Page.Text) = 0 then
  begin
    AnalyzeMoviePage(Page)
  end
  else
  begin
    if pos('Nici un rezultat', Page.Text) > 0 then
      ShowMessage('Sorry, no results were found for ' + MovieName)
    else
    begin
      PickTreeClear;
      AddMoviesTitles(Page);
      if PickTreeExec(Address) then
        AnalyzePage(Address);
    end
  end;
  Page.Free;
end;

function FindValue(BeginTag, EndTag: string; Page: TStringList; var LineNr: Integer; var Line: string): string;
var
  BeginPos, EndPos: Integer;
  Value: string;
begin
  Result := '';
  Value := '';
  BeginPos := Pos(BeginTag, Line);
  if BeginPos > 0 then
  begin
    BeginPos := BeginPos + Length(BeginTag);
    if BeginTag = EndTag then
    begin
      Delete(Line,1,BeginPos-1);
      BeginPos := 1;
    end;
    EndPos := pos(EndTag, Line);
    while ((EndPos = 0) and (LineNr < Page.Count-1 )) do
    begin
      Value := Value + copy(Line, BeginPos, Length(Line) - BeginPos);
      // Next Line
      BeginPos := 1;
      LineNr := LineNr + 1;
      Line := Page.GetString(LineNr);
      if Value = '' then
        Exit;
      EndPos := Pos(EndTag, Line);
    end;
    Value := Value + copy(Line, BeginPos, EndPos - BeginPos);
   end;
  Result := Value;
end;

procedure AnalyzeMoviePage(Page: TStringList);
var
  Line, Value, Value2: string;
  IsFirst, LineNr, BeginPos, EndPos, DescrImport: Integer;
  HasPressReviews : Integer;
  PressReviewAddress : string;
begin
  LineNr := FindLine('<h1 class="movie">', Page, 0);
  if LineNr > 2 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('<a href="', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 9;
    EndPos := pos('style=', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, BeginPos, EndPos - BeginPos - 2);
    HTMLDecode(Value);
    MovieURL := 'http://www.cinemagia.ro' + Value;

    // URL
    SetField(fieldURL, MovieURL);
   
    Value := FindValue( '<b>', '</b>', Page, LineNr, Line);
    HTMLDecode(Value);
    SetField(fieldTranslatedTitle, Value);

    Value := FindValue( '<i>(', ')</i>', Page, LineNr, Line);
    HTMLDecode(Value);
    SetField(fieldYear, Value);

    LineNr := LineNr + 1;
    Line := Page.GetString(LineNr);

    Value := FindValue( '<i>', '</i>', Page, LineNr, Line);
    HTMLDecode(Value);
    SetField(fieldOriginalTitle, Value);
  end;

  LineNr := FindLine('<img src="/getimg.php?id=', Page, LineNr);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('<img src="/getimg.php?id=', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 10;
    EndPos := pos('&size=s"', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, BeginPos, EndPos - BeginPos + 6) + 'm';
    HTMLDecode(Value);

    GetMoviePicture( 'http://www.cinemagia.ro' + Value);
  end;
 
  LineNr := FindLine('Regia<br>', Page, LineNr);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('Regia<br>', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 9;
     
    Delete( Line, 1, BeginPos);
    EndPos := pos('<br>', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, 1, EndPos);

    BeginPos := pos('">', Value) + 2;
    EndPos := pos('</a>', Value);
    Value := copy(Value, BeginPos, EndPos - BeginPos);

    HTMLDecode(Value);
    SetField(fieldDirector, Value);
  end;

  LineNr := FindLine('Gen<br>', Page, LineNr);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('Gen<br>', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 7;

    Delete( Line, 1, BeginPos);
    EndPos := pos('<br>', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, 1, EndPos);

    Value2 := '';
    repeat
      BeginPos := pos('">', Value) + 2;
      EndPos := pos('</span>', Value);
      Value2 := Value2 + ', ' + copy(Value, BeginPos, EndPos - BeginPos);
      Delete( Value, 1, EndPos + 8);
    until pos('">', Value) < 1;
   
    Delete( Value2, 1, 2);

    HTMLDecode(Value2);
    SetField(fieldCategory, Value2);
  end;
 
  LineNr := FindLine('what=cast', Page, LineNr);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('<a href="', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 9;

    EndPos := pos('" class', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, BeginPos, EndPos - BeginPos);

    SetField(fieldActors, GetCast('http://www.cinemagia.ro' + Value));
  end;

  LineNr := FindLine('Durata<br>', Page, LineNr);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('Durata<br>', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 10;

    Delete( Line, 1, BeginPos);
    EndPos := pos('</td>', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, 1, EndPos);

    BeginPos := pos('">', Value) + 2;
    EndPos := pos(' minute</span>', Value);
    Value := copy(Value, BeginPos, EndPos - BeginPos);

    HTMLDecode(Value);
    SetField(fieldLength, Value);
  end;
  
  LineNr := FindLine('Produs de<br>', Page, LineNr);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('Produs de<br>', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 12;

    Delete( Line, 1, BeginPos);
    EndPos := pos('</td>', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, 1, EndPos);

    BeginPos := pos('">', Value) + 2;
    EndPos := pos('</span>', Value);
    Value := copy(Value, BeginPos, EndPos - BeginPos);

    HTMLDecode(Value);
    SetField(fieldProducer, Value);
  end;

  HasPressReviews := 0;
  IsFirst := 1;
  Value2 := '';
  LineNr := FindLine('Continuare</a>', Page, 0);
  while (LineNr > -1) do
  begin
    Line := Page.GetString(LineNr);
    BeginPos := pos('<a href="/movie.php', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 9;
     
    EndPos := pos('&hist=0', Line);
    if EndPos = 0 then
      EndPos := Length(Line);
    Value := copy(Line, BeginPos, EndPos - BeginPos + 7);
    HTMLDecode(Value);

    if pos( 'what=pressreviews', Value) > 0 then
    begin
      PressReviewAddress := Value;
      HasPressReviews := 1;
    end
    else
    begin
      if IsFirst = 1 then
      begin
        SetField(fieldDescription, GetDescriptions( Value));
        IsFirst := 0;
      end
      else
      begin
        Value2 := Value2 + GetDescriptions( Value) + '<br/>';
      end;
    end;

    LineNr := FindLine('Continuare</a>', Page, LineNr + 1);
  end;
 
  if HasPressReviews = 1 then
  begin
    Value2 := Value2 + GetPressReviews( PressReviewAddress);
  end;
 
  SetField(fieldComments, Value2);

  //DisplayResults;
end;

procedure GetMoviePicture( PictureAddress: string);
begin
   GetPicture(PictureAddress);
end;

function GetCast(Address: string): string;
var
  Line, Value: string;
  LineNr: Integer;
  NoInfiniteLoop : Integer;
  BeginPos, EndPos: Integer;
  Page: TStringList;
begin
  Result := '';
  Page := TStringList.Create;
  Page.Text := GetPage(Address);
  LineNr := FindLine('Distributie', Page, 0);
  Value := '';
  if LineNr > -1 then
  begin
    LineNr := LineNr + 2;
    Line := Page.GetString(LineNr);
   
    NoInfiniteLoop := 0;
    while ( pos('</table>', Line) < 1 ) and ( NoInfiniteLoop < 10) do
    begin
      NoInfiniteLoop := NoInfiniteLoop + 1;
      Line := Page.GetString(LineNr + 1);
      BeginPos := pos('">', Line);
      if BeginPos > 0 then
        BeginPos := BeginPos + 2;
      EndPos := pos('</a>', Line);
      if EndPos < 1 then
        EndPos := Length(Line);
      if Value <> '' then
        Value := Value + '; ';
      Value := Value + copy(Line, BeginPos, EndPos - BeginPos);

      Line := Page.GetString(LineNr + 2);
      BeginPos := pos('<td>', Line);
      if BeginPos > 0 then
        BeginPos := BeginPos + 4;
      EndPos := pos('</td>', Line);
      if EndPos < 1 then
        EndPos := Length(Line);
      if Value <> '' then
        Value := Value + ' ';
      Value := Value + '(in rolul ' + copy(Line, BeginPos, EndPos - BeginPos) + ')';
   
      Line := Page.GetString(LineNr + 4);
      LineNr := LineNr + 4;
    end;

    HTMLDecode(Value);

    Result := Value;
  end;
  Page.Free;
end;

function GetDescriptions(Address: string): string;
var
  Line, Value: string;
  LineNr: Integer;
  BeginPos, EndPos: Integer;
  Page: TStringList;
begin
  Result := '';
  Page := TStringList.Create;
  Page.Text := GetPage(Address);
  LineNr := FindLine('Pagina principala', Page, 0);
  if LineNr > -1 then
  begin
    Line := Page.GetString(LineNr + 1);
    BeginPos := pos('<div class="section_title">', Line);
    if BeginPos > 0 then
      BeginPos := BeginPos + 27;
    EndPos := pos('</div>', Line);
    if EndPos < 1 then
      EndPos := Length(Line);

    Value := copy(Line, BeginPos, EndPos - BeginPos);
    HTMLRemoveTags(Value);
    Value := '<b>' + Value + '</b><br/>';

    LineNr := LineNr + 2;
    repeat
      Line := Page.GetString(LineNr);
      BeginPos := pos('<div class="section">', Line);
      if BeginPos > 0 then
        BeginPos := BeginPos + 21
      else
        BeginPos := 1;
      EndPos := pos('</div>', Line);
      if EndPos < 1 then
        EndPos := Length(Line) + 1;
      if Value <> '' then
        Value := Value + ' ';
      Value := Value + copy(Line, BeginPos, EndPos - BeginPos);
      LineNr := LineNr + 1;
    until (pos('</div>', Line) > 0) or (LineNr = Page.Count);
    HTMLDecode(Value);

    Result := Value;
  end;

  Page.Free;
end;

function GetPressReviews(Address: string): string;
var
  Line, Value, Value2: string;
  LineNr: Integer;
  BeginPos, EndPos: Integer;
  Page: TStringList;
begin
  Result := '';
  Value2 := '';
  Value := '';
  Page := TStringList.Create;
  Page.Text := GetPage(Address);
  LineNr := FindLine('<div class="section_title">', Page, 0);
  while LineNr > -1 do
  begin
    Line := Page.GetString(LineNr);
    if pos('a href=', Line) < 1 then
    begin
      BeginPos := pos('<div class="section_title">', Line);
      if BeginPos > 0 then
        BeginPos := BeginPos + 27;
      EndPos := pos('</div>', Line);
      if EndPos < 1 then
        EndPos := Length(Line);

      Value := copy(Line, BeginPos, EndPos - BeginPos);
      HTMLRemoveTags(Value);
      Value := '<b>' + Value + '</b><br/>';

      LineNr := LineNr + 2;
      repeat
        Line := Page.GetString(LineNr);
        BeginPos := pos('<div class="section">', Line);
        if BeginPos > 0 then
          BeginPos := BeginPos + 21
        else
          BeginPos := 1;
        EndPos := pos('</div>', Line);
        if EndPos < 1 then
          EndPos := Length(Line) + 1;
        if Value <> '' then
          Value := Value + ' ';
        Value := Value + copy(Line, BeginPos, EndPos - BeginPos);
        LineNr := LineNr + 1;
      until (pos('</div>', Line) > 0) or (LineNr = Page.Count);
   
      Value2 := Value2 + Value;
    end
    else
      LineNr := LineNr + 1;

    LineNr := FindLine('<div class="section_title">', Page, LineNr);
  end;
 
  HTMLDecode(Value2);

  Result := Value2;

  Page.Free;
end;

procedure AddMoviesTitles(Page: TStringList);
var
  Line: string;
  LineNr: Integer;
  MovieTitle, MovieAddress: string;
  StartPos: Integer;
begin
  LineNr := FindLine( 'Rezultatele cautarii', Page, 0);
  if LineNr > -1 then
  begin
    PickTreeAdd('Filme', '');
    LineNr := LineNr + 3;
    Line := Page.GetString(LineNr);
    repeat
      StartPos := pos('href="', Line) + 5;
      Delete(Line, 1, StartPos);
      MovieAddress := Copy(Line, 1, pos('"', Line) - 1);
      StartPos := Pos('">', Line) + 2;
      MovieTitle := Copy(Line, StartPos, Pos('</i>', Line) - StartPos);
      HTMLDecode(Movietitle);
      Delete( Movietitle, Pos('</a>', Movietitle), 4);
      Delete( Movietitle, Pos('<i>', Movietitle), 3);
      PickTreeAdd(MovieTitle, 'http://www.cinemagia.ro' + MovieAddress);
      LineNr := LineNr + 1;
      Line := Page.GetString(LineNr);
    until Pos('</div>', Line) > 0;
  end;
end;

begin
  if CheckVersion(3,5,0) then
  begin
    MovieName := GetField(fieldOriginalTitle);
    if MovieName = '' then
      MovieName := GetField(fieldTranslatedTitle);
    if Input('Cinemagia.RO Import', 'Enter the title of the movie:', MovieName) then
    begin
      AnalyzePage('http://www.cinemagia.ro/search.php?q=' + UrlEncode(MovieName) + '&hist=0');
    end;
  end
  else
  ShowMessage('This script requires a newer version of Ant Movie Catalog (at least the version 3.5.0)');
end.

Enjoy

Post by **antp** » 2005-06-16 13:39:51

Thanks