[Randall Lind]

No maybe it didn't rename it for the older version. I did delete every thing but for jobslog.emm and media.emm and I re-scrape and that is what I got. Once I get done I will add some new movies and scrape those.

[ynksbsbll2]

Hi All,

So if I have a previously working version of EMM how exactly do I handle 'upgrading' to this new 'Ember 1.3.0.7 Fix1' version?

I'm sure the details are in this thread somewhere, but there seems to several versions floating around...this thread size is getting on the largish size.

Are there any default XML files I should preserve / copy or can I simply reinstall new EMM version to a different directy?

TIA

I just copy everything from the new version to my current install and choose "Copy and Replace" on any duplicate files.

[Randall Lind]

Dan what's different? It is still adding strange titles on auto. Example: Race to Witch Mountain is Leo Little's Big Show" Race to Witch Mountain. Manually also looks the same to me as well. I base this off of 1.3.0.7a from DarkKnight which I deleted and replace everything but for jobslogs.emm and media.emm IMDB on both his and fix1 is the same from what I see.

Now I haven't tried Youtube yet

You sure you haven't got builds mixed up because I just tested "Race to Witch Mountain" on auto and got exactly that returned.

Did you delete the nfo from the other builds giving wrong results ?

I did delete the nfo of some files like you said and got the same weird names so I guess I will delete all files including .emm and report back

[Randall Lind]

<<X>>, seem to be working good now what was strange after I deleted all files and then unrar emmber fix 1. I setup my directory most movies only listed posters. I just got done scraping them all. So her I go again LOL

Auto scrape is working great now. Thanks for the help

[thethirdnut]

@all - thanks for re-installing tips - I'll try it out tomorrow / Fri evening over a beer or two.

With this new install as well - is there specific settings / best-practice for scraper, fanart, etc settings? I have OLD data on this somewhere, but am guessing this likely changed with the newer versions as well.

I could be convinced to properly document and reproduce this if I knew where to find it.

[METTTAL]

Anyone else getting virus messages after extracting the new version? Norton threw me all kinds of threat warnings that never appeared on my original Ember folder.

[DarkKnight]

Dan what's different? It is still adding strange titles on auto. Example: Race to Witch Mountain is Leo Little's Big Show" Race to Witch Mountain. Manually also looks the same to me as well. I base this off of 1.3.0.7a from DarkKnight which I deleted and replace everything but for jobslogs.emm and media.emm IMDB on both his and fix1 is the same from what I see.

Now I haven't tried Youtube yet

You sure you haven't got builds mixed up because I just tested "Race to Witch Mountain" on auto and got exactly that returned.

Did you delete the nfo from the other builds giving wrong results ?

I did delete the nfo of some files like you said and got the same weird names so I guess I will delete all files including .emm and report back

Even in 1308, some of the movies scrape fine on auto, and some choose the wrong one. Although Dan has changed the http request to imdb somewhat, I'm not seeing a difference in the end result. Resident evil scrapes fine, Ice-age chose 'Projector" Ice-Age'

Also, I don't really see why someone would upload their release to a "fremium" host when Github hosts the files free without issue. I don't like dealing with ad clogged pay-per-click services, so I uploaded Dan's version to my github repo: https://github.com/downloads/Darkknight3...7_fix1.rar

Maybe when Dan posts his source, we might make some further progress tuning the auto movie scrape process.

[FreddieP786]

Hey there everyone, new version seems to work fine on movies. But I cant seem to scrape any TV series.

I deleted the old version completely and installed the fix as a new version. Was that wrong?

Is there any way to fix it?

[rodercot]

2nd - Not scraping TV shows properly. I mentioned this with the 1308 fix as well.

At least with the 1307 fix version I can see and change the tv sources under setting - tvshows. but it does not scrape properly on any TV shows. I am still using 1307 and adding manually adding the imdb selection. I just test the new version as a fresh install and let it rebuild the library from scratch.

THX,

Dave

[FreddieP786]

I tried other alternative software, don't like any of them. Ember has been a favorite for a long time now. I really hope there is a fix soon. I don't mind waiting for one.

EDIT: Thank you so much for all the brilliant work so far. You guys are amazing and your efforts don't go unnoticed or unappreciated.

[PogMoThoin]

My Ember has stopped grabbing fan art, nfo's work fine, just the fan art. I've updated to 1307 fix1 but it's not fixed it. Is it the same issue you all are having?

[DanCooper]

Even in 1308, some of the movies scrape fine on auto, and some choose the wrong one. Although Dan has changed the http request to imdb somewhat, I'm not seeing a difference in the end result. Resident evil scrapes fine, Ice-age chose 'Projector" Ice-Age'

Also, I don't really see why someone would upload their release to a "fremium" host when Github hosts the files free without issue. I don't like dealing with ad clogged pay-per-click services, so I uploaded Dan's version to my github repo: https://github.com/downloads/Darkknight3...7_fix1.rar

Maybe when Dan posts his source, we might make some further progress tuning the auto movie scrape process.

For me it works perfectly (tested with autoscraping).
ice-age.avi = Ice Age (2002)
Race to Witch Mountain.avi = Race to Witch Mountain (2009)

I did not know that you also can host files on Github.

2nd - Not scraping TV shows properly. I mentioned this with the 1308 fix as well.

At least with the 1307 fix version I can see and change the tv sources under setting - tvshows. but it does not scrape properly on any TV shows. I am still using 1307 and adding manually adding the imdb selection. I just test the new version as a fresh install and let it rebuild the library from scratch.

THX,

Dave

TV shows scraping also works without problems.
If you've reinstalled Ember you need rescrape ANY tv-show once again, so Ember rebuilds his database.


I made the following changes in the source code:

Code:

Private Const TABLE_PATTERN As String = "<table.*?>\n?(.*?)</table>"

.........

        Private Function SearchMovie(ByVal sMovie As String) As MovieSearchResults
            Try

                Dim D, W As Integer
                Dim R As New MovieSearchResults

                Dim sHTTP As New HTTP
                Dim HTML As String = sHTTP.DownloadData(String.Concat("http://", IMDBURL, "/find?q=", Web.HttpUtility.UrlEncode(sMovie, System.Text.Encoding.GetEncoding("ISO-8859-1")), "&s=all"))
                Dim HTMLm As String = sHTTP.DownloadData(String.Concat("http://", IMDBURL, "/find?q=", Web.HttpUtility.UrlEncode(sMovie, System.Text.Encoding.GetEncoding("ISO-8859-1")), "&s=tt&ttype=ft&ref_=fn_ft"))
                Dim HTMLe As String = sHTTP.DownloadData(String.Concat("http://", IMDBURL, "/find?q=", Web.HttpUtility.UrlEncode(sMovie, System.Text.Encoding.GetEncoding("ISO-8859-1")), "&s=tt&ttype=ft&exact=true&ref_=fn_tt_ex"))
                Dim rUri As String = sHTTP.ResponseUri
                sHTTP = Nothing

                'Check if we've been redirected straight to the movie page
                If Regex.IsMatch(rUri, IMDB_ID_REGEX) Then
                    Dim lNewMovie As MediaContainers.Movie = New MediaContainers.Movie(Regex.Match(rUri, IMDB_ID_REGEX).ToString, _
                        StringUtils.ProperCase(sMovie), Regex.Match(Regex.Match(HTML, MOVIE_TITLE_PATTERN).ToString, "(?<=\()\d+(?=.*\))").ToString, 0)
                    R.ExactMatches.Add(lNewMovie)
                    Return R
                End If

                'D = HTML.IndexOf("<b>Popular Titles</b>")
                D = HTML.IndexOf("</a>Titles</h3>")
                If D <= 0 Then GoTo mPartial
                W = HTML.IndexOf("</table>", D) + 8

                Dim Table As String = Regex.Match(HTML.Substring(D, W - D), TABLE_PATTERN).ToString

                Dim qPopular = From Mtr In Regex.Matches(Table, TITLE_PATTERN) _
                               Where Not DirectCast(Mtr, Match).Groups("name").ToString.Contains("<img") AndAlso Not DirectCast(Mtr, Match).Groups("type").ToString.Contains("VG") _
                               Select New MediaContainers.Movie(GetMovieID(DirectCast(Mtr, Match).Groups("url").ToString), _
                                                Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString), Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("year").ToString), StringUtils.ComputeLevenshtein(StringUtils.FilterYear(sMovie).ToLower, StringUtils.FilterYear(Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString)).ToLower))

                R.PopularTitles = qPopular.ToList
mPartial:

                'D = HTML.IndexOf("Titles (Partial Matches)")
                D = HTMLm.IndexOf("</a>Titles</h3>")
                If D <= 0 Then GoTo mApprox
                W = HTMLm.IndexOf("</table>", D) + 8

                Table = Regex.Match(HTMLm.Substring(D, W - D), TABLE_PATTERN).ToString
                Dim qpartial = From Mtr In Regex.Matches(Table, TITLE_PATTERN) _
                    Where Not DirectCast(Mtr, Match).Groups("name").ToString.Contains("<img") AndAlso Not DirectCast(Mtr, Match).Groups("type").ToString.Contains("VG") _
                    Select New MediaContainers.Movie(GetMovieID(DirectCast(Mtr, Match).Groups("url").ToString), _
                                     Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString), Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("year").ToString), StringUtils.ComputeLevenshtein(StringUtils.FilterYear(sMovie).ToLower, StringUtils.FilterYear(Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString)).ToLower))

                R.PartialMatches = qpartial.ToList
mApprox:

                'Now process "Approx Matches" and merge both Partial and Approx matches
                D = HTML.IndexOf("Titles (Approx Matches)")
                If D <= 0 Then GoTo mExact
                W = HTML.IndexOf("</table>", D) + 8

                Table = Regex.Match(HTML.Substring(D, W - D), TABLE_PATTERN).ToString

                Dim qApprox = From Mtr In Regex.Matches(Table, TITLE_PATTERN) _
                    Where Not DirectCast(Mtr, Match).Groups("name").ToString.Contains("<img") AndAlso Not DirectCast(Mtr, Match).Groups("type").ToString.Contains("VG") _
                    Select New MediaContainers.Movie(GetMovieID(DirectCast(Mtr, Match).Groups("url").ToString), _
                                     Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString), Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("year").ToString), StringUtils.ComputeLevenshtein(StringUtils.FilterYear(sMovie).ToLower, StringUtils.FilterYear(Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString)).ToLower))

                If Not IsNothing(R.PartialMatches) Then
                    R.PartialMatches = R.PartialMatches.Union(qApprox.ToList).ToList
                Else
                    R.PartialMatches = qApprox.ToList
                End If

mExact:

                'D = HTML.IndexOf("Titles (Exact Matches)")
                D = HTMLe.IndexOf("</a>Titles</h3>")
                If D <= 0 Then GoTo mResult
                W = HTMLe.IndexOf("</table>", D) + 8

                Table = String.Empty
                Table = Regex.Match(HTMLe.Substring(D, W - D), TABLE_PATTERN).ToString

                Dim qExact = From Mtr In Regex.Matches(Table, TITLE_PATTERN) _
                               Where Not DirectCast(Mtr, Match).Groups("name").ToString.Contains("<img") AndAlso Not DirectCast(Mtr, Match).Groups("type").ToString.Contains("VG") _
                               Select New MediaContainers.Movie(GetMovieID(DirectCast(Mtr, Match).Groups("url").ToString), _
                            Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString.ToString), Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("year").ToString), StringUtils.ComputeLevenshtein(StringUtils.FilterYear(sMovie).ToLower, StringUtils.FilterYear(Web.HttpUtility.HtmlDecode(DirectCast(Mtr, Match).Groups("name").ToString)).ToLower))

                R.ExactMatches = qExact.ToList

mResult:
                Return R
            Catch ex As Exception
                Master.eLog.WriteToErrorLog(ex.Message, ex.StackTrace, "Error")
                Return Nothing
            End Try
        End Function

For popular matches:
http://akas.imdb.com/find?q=ice+age&s=all

For more/partial matches:
http://akas.imdb.com/find?q=ice+age&s=tt...ref_=fn_ft

For exact matches:
http://akas.imdb.com/find?q=ice+age&s=tt..._=fn_tt_ex


Currently this is the only way to get useful results from IMDB.

[DanCooper]

My Ember has stopped grabbing fan art, nfo's work fine, just the fan art. I've updated to 1307 fix1 but it's not fixed it. Is it the same issue you all are having?

As always, no problems for me
Can you give me an example?

---

Ember 1.3.0.7 Fix1:

• IMDB fix created, developed and presented by me, myself and i
  

Would you mind posting that fix here in the thread? I'm interested in what's changed.

Github is updated...

[DanCooper]

Seems to work great so far

Only one strange error which I had with the others to

Code:

Message: Access to the path 'C:\Program Files\Ember\AdvancedSettings.xml' is denied.

Seems like a false error in any case as AdvancedSettings.xml does actually update with any new changes

I have the problem sometimes also. I think Ember is sometimes too fast when editing and blocked himself.

[DanCooper]

Where can I even get german outlines? When I used german outline in Ember it was just the first half of the plot for most movies.

At the OFDB not possible to convert.

OFDB verwendet eine gewisse Anzahl an Zeichnen aus dem Plot als Outline. Es gibt also eigentlich keine richtige/saubere Outline die gescrapt werden könnte.